初创公司尽调
尽调报告 healthcare / biotech Series E / late-stage private 2026-05-20

Aidoc

医疗 AI 尽调报告

Aidoc 是可信、已有规模的临床 AI 平台,但公开证据对价格和经济性仍过于不透明,最多只支持继续研究。

封面要素

最新披露轮次 01
150 USD M [CV003]
已披露累计融资 02
Over 500 USD M [CO021]
部署覆盖 03
Nearly 2,000 hospitals [CO025]
投资建议 04
research-more [CV043]

公司概况

Aidoc 是一家私营临床 AI 公司,由 Elad Walach、Michael Braginsky 和 Guy Reiner 于 2016 年创立。公司销售 Aidoc aiOS,这是一个企业级临床 AI 操作系统,把放射科分诊、照护协同和工作流治理嵌入医院 PACS 与 EHR 环境,并在其上叠加 CARE 基础模型能力。2025-2026 年的公开证据显示,Aidoc 资本充足、部署规模已成形,在医疗系统中覆盖广泛,FDA / CE 进展也在持续推进;但当前估值、收入质量、准确员工数,以及常被引用的 170+ 模型数量仍未披露。

官网
www.aidoc.com
成立时间
2016-01-01
创始人
Elad Walach, Michael Braginsky, Guy Reiner
创立地点
Tel Aviv, Israel
总部
New York, NY
产品
企业级临床 AI 平台,结合 Aidoc aiOS 编排、FDA 批准的放射科解决方案、照护协同工作流和 CARE 基础模型能力,用于多病种分诊和下游行动。
客户
需要影像分诊、照护协同和企业级 AI 治理的医疗系统、放射科团队和急症照护服务线。
商业模式
面向医院和医疗系统销售的企业 SaaS / 站点许可型临床 AI 软件,核心是工作流集成、编排和照护协同模块,而不是自助式交易定价。
阶段
Series E / late-stage private
融资情况
2026 年 4 月 Series E 轮融资 $150M,由 Goldman Sachs Alternatives 的 Growth Equity 领投;官方材料称累计融资现已超过 $500M,但当前投后估值仍未披露。
[CO001, CO002, CO003, CO004, CO005, CO019, CO021, CO025]

执行摘要

主要优势

  • 披露累计融资超过 $500M,且后期投资人支持延续到 2026 年 Series E。
  • 部署口径已接近 2,000 家医院,并有 Asklepios、Hartford HealthCare 等具名企业级落地。
  • aiOS 加 CARE 把 Aidoc 从单用途放射科算法供应商,推向企业级编排和工作流平台。
  • 持续推进 FDA 和 CE,降低了 Aidoc 只是研究阶段 AI 故事的风险。

主要风险

  • 抓取到的 2024-05-20 之后来源没有清楚披露当前投后估值,因此独角兽身份和进入价格仍未验证。
  • ARR、收入、毛利率、留存和现金 / 烧钱速度均未披露,无法真正承销软件质量。
  • 报销和采购摩擦可能让医院继续把 Aidoc 当作有用的工作流层,而不是高溢价独立平台。
  • 公开规模披露混用医院、医疗系统、患者病例和支持患者数,限制了对收入的直接推断。

未决问题

  • 2025-2026 年融资的当前投后估值、轮次条款和优先权堆叠。
  • ARR、按产品线拆分的收入、毛利率、NRR、CAC / 回本周期、现金余额和循环信贷使用。
  • 稳定的当前 FDA 批准数量,以及有来源支持的当前产品组合 / 模型数量。
  • 精确公开员工数,以及客户账户集中度 / 扩张数据。

目录

Chapter 01

01公司概况

1.1 身份、覆盖与产品架构

Aidoc 的公开材料一直把公司定位为临床 AI 厂商,而不是狭窄的放射科算法公司。官方 about 页面称,公司专注于帮助医疗团队优化患者治疗和经济价值;aiOS 与 CARE 页面则显示,这一定位已经从影像分诊扩展到更宽的企业级工作流层。aiOS 被描述为控制平面,负责在医疗系统内运行、编排和治理临床 AI,连接 PACS、EHR、移动端和照护工具。CARE 则被定位为多模态基础模型,让 Aidoc 从单病种工具走向更广的疾病覆盖、自动测量,并最终生成报告草稿。这一点重要,因为它改变了公司的投资逻辑:Aidoc 不再只向放射科医生销售急性发现提醒,而是在销售一个面向多算法部署、照护协同和模型治理的运行层。获取的来源支持其通过 2026 年融资公告形成了清晰的纽约商业姿态,而 Guy Reiner 的官方简介仍明确称其为特拉维夫分部总经理。由此看到的是一家技术根基在以色列、资本市场形象面向纽约、战略围绕企业级部署而非科室级单点方案展开的公司。[CO001, CO006, CO007, CO008, CO009, CO010]

FO002: 公司快照逻辑

Aidoc 的创始人、CARE 模型、aiOS 平台和企业部署如何连接起来,拼出当前公司投资逻辑。

[CO002, CO005, CO009, CO011, CO013, CO019]

1.2 创始人、领导层与关键人物集中度

对一家私营公司而言,Aidoc 的创始团队得到的公开佐证异常充分。官方领导层简介确认 Elad Walach 为联合创始人兼 CEO,Michael Braginsky 为联合创始人兼 CTO,Guy Reiner 为联合创始人、首席架构师和特拉维夫分部总经理。CTech 2025 年 7 月的报道与这些角色一致,并将公司创立时间追溯到 2016 年。从职能上看,三人仍在支撑核心商业模式:Walach 是对外发言人和融资声音,Braginsky 在技术与融资报道中都以基础模型架构师身份出现,Reiner 对特拉维夫分部的职责则意味着产品和工程执行有很强连续性。公开可见的非创始高管确实存在,但披露远不如创始人完整;本次检索到的记录对创始人领导力最强,对更广泛管理梯队则薄得多。因此,关键人物集中度是一个真实的尽调项,而不是泛泛的创业公司风险提示。Aidoc 的公开材料确实显示组织成熟度在提升,例如明确讨论治理、漂移检测和企业级监控;但公开记录仍未提供完整领导层名单,也无法清楚展示创始人以下的授权分布如何随版本迭代变化。[CO002, CO003, CO004, CO005, CO034, CO035]

领导层与创始人表
人物职位背景创始人-市场匹配 / 职能覆盖关键人依赖
Elad Walach共同创始人兼 CEO融资、产品愿景和客户扩张的公司公开声音主导市场叙事、融资和企业销售落地框架高——主要公开发言人和战略承载者
Michael Braginsky共同创始人兼 CTOCARE 基础模型、路线图和模型质量论证的技术负责人连接产品架构、FDA 证据和模型开发经济性高——基础模型可信度集中在此
Guy Reiner共同创始人、首席架构师、特拉维夫分支 GM负责架构和以色列技术分支;历史上主导 CE/FDA 批准工作稳住以色列研发连续性和算法发布节奏中高——分支运营与产品执行之间的关键连接
Reut Yalon首席产品官在 2025 年基础模型许可公告中作为产品负责人被引用显示创始人之外的产品管理层正在扩展中——可见度较高,但引用频率低于创始人
Andy Crowder首席数字官在 aiOS 平台页面署名,并与企业部署叙事相关代表工作流、数字化转型和运营平台信息中——有用的企业级声音,但不是技术逻辑核心

该公开名单有意只覆盖一部分:它抓取官方简介和近期公告中证据最充分的领导者,不是完整高管名录。

[CO002, CO003, CO004, CO005, CO034, CO035]

1.3 融资历史、投资人组合与估值可见度

Aidoc 的资本路径在过去两年明显提速。官方 2022 年 Series D 公告记录了一轮 $110 million 融资,使累计融资达到 $250 million。到 2025 年 7 月,PR Newswire、Calcalist 和 Globes 均报道了由 General Catalyst 与 Square Peg 领投的新一轮 $150 million 成长融资,NVentures 和数家美国大型医疗系统也参与;PR Newswire 和 Globes 还提到额外 $40 million 循环信贷额度,以及 $370 million 的累计融资总额。随后,2026 年 4 月,Aidoc 和 Goldman Sachs Alternatives 分别宣布由 Goldman 领投的 $150 million Series E,General Catalyst、SoftBank Vision Fund 2 和 NVentures 跟投。2026 年 4 月的材料只称总融资现为“超过 $500 million”,没有给出精确数字。这一区别很重要:它足以证明公司资本化程度很高,却不足以解决估值问题。该点上最强的 2024 年后来源其实是 Globes 2025 年 7 月的采访,Michael Braginsky 明确表示公司不会披露该轮融资估值。因此,在本次报告中,公司是否为私营独角兽仍是实质证据缺口:高融资和规模提升都很清楚,但未获取到 2024-05-20 之后披露估值、足以确认当前独角兽标记的来源。[CO015, CO016, CO017, CO018, CO019, CO020]

利益相关方 / 投资人图谱
利益相关方角色控制权或经济重要性尽调问题
Goldman Sachs Alternatives2026 年 Series E 领投方最新领投机构投资人;当前轮次中最强外部背书确认董事会权利、清算优先权,以及估值上调是否实质性
General Catalyst2025 年轮次的领投或重复成长投资人;2026 年 Series E 参与方连续融资中出现,因此可能是核心关系投资人确认持股集中度及任何治理影响
Square Peg共同领投 2025 年 CARE 融资基础模型扩展逻辑的关键支持者确认其在 2026 年虽未列名领投,是否仍按比例跟投
NVentures2025 年和 2026 年轮次参与方可在资本之外发挥作用的战略 GPU/AI 生态投资人验证排他性、算力抵扣或商业化义务
SoftBank Vision Fund 22026 年 Series E 参与方会影响未来融资选择权的后期资本伙伴确认持仓规模和后续跟投预期
TCV共同领投 2022 年 Series D与 Aidoc 平台扩张阶段相关的早期成长阶段支持者检查 2025-2026 年稀释后的持股
Alpha Intelligence Capital共同领投 2022 年 Series D较早的 AI 专业投资人;体现技术投资人背书澄清当前持股及董事会或观察员角色
战略医疗系统投资者Hartford、Mercy、Sutter、WellSpan 在 2025 年融资语境中被点名运营上重要,因为客户资本会影响部署路径确认每位投资者是否也是企业客户,以及条款如何

该图谱覆盖 2022、2025 和 2026 年轮次公开披露的主要财务和战略利益相关方。它不是股权结构表,也不揭示持股比例或治理权。

[CO016, CO018, CO020, CO041]

1.4 规模信号、具名客户与监管里程碑

Aidoc 公开披露的规模已经足够大,分析公司时应把它视为企业级部署故事,而不是试点阶段厂商。2025 年 CARE 融资稿称,Aidoc 每年在 150+ 个医疗系统中支持超过 45 million 名患者,并预计三年内达到 100 million。2026 年 Series E 材料进一步提高了披露规模,称年度患者病例超过 60 million,累计已分析患者病例超过 110 million,并部署在近 2,000 家医院。这些数字来自公司自报,但支撑它们的客户证据已强于早期轮次。Asklepios 披露了覆盖德国 28 家医院的部署,每月约分析 35,000 张 CT 和 X-ray 图像;Hartford HealthCare 则表示 Aidoc 的 17 个算法部署在三周内上线,覆盖数百万次患者检查。监管记录也很具体,而不是纯叙事。FDA K213721 显示 2022 年脑动脉瘤分诊获批,K231631 显示 2023 年 CAC 量化获批。Aidoc 2025 年和 2026 年新闻稿随后从逐适应症批准转向基础模型叙事:先是 CARE1 的肋骨骨折获批,再到综合体部 CT 分诊获批,把 11 个新适应症与 3 个既有适应症纳入同一工作流。客户证据和监管产出的组合,是后续章节可以把 Aidoc 视为已成规模临床 AI 基础设施公司,而不是单产品创业公司的核心原因。[CO022, CO023, CO024, CO025, CO026, CO027]

KPI 快照表
指标数值 / 状态日期 / 时间口径置信度缺口 / 备注
成立2016历史Calcalist 和 Fierce Healthcare 交叉印证
创始人3 位已验证共同创始人当前Elad Walach、Michael Braginsky、Guy Reiner 等创始人
最新披露轮次Series E 轮 $150M2026-04由 Goldman Sachs Alternatives 领投
已披露累计融资超过 $500M2026-042026 年官方轮次材料已不再列出精确总额
较早累计融资检查点$370M2025-07包括 $150M 股权融资和 $40M 循环信贷额度
当前披露覆盖范围近 2,000 家医院2026-04公司与投资人新闻稿在该表述上吻合
年度规模指标每年分析 60M+ 个患者病例2026-042026 年官方轮次材料
支持患者数每年 ~70M 名患者2026-04投资人侧新闻稿表述;公司侧文案强调病例
医疗系统数量150+ 家医疗系统2025-07较早但仍属近期的公司说法
当前估值 / 独角兽状态未确认2026抓取到的 2024-05-20 之后来源均未披露估值
精确公开员工数未确认2026抓取来源未披露精确员工数
170+ AI 模型数量未验证2026抓取到的官方来源支持广泛算法覆盖,但不支持这一数量

官方来源支持当前融资和部署规模,但估值、员工数和常被引用的 170+ 模型数量仍未验证。与患者有关的指标混用“病例”和“患者”,因此本表保留最接近各来源的措辞。

[CO001, CO015, CO017, CO019, CO021, CO022]
里程碑表
日期事件类型金额 / 估值 / 状态参与方含义
2016Aidoc 成立创立已成立Elad Walach、Michael Braginsky、Guy Reiner 等创始人确立一家源自以色列的临床 AI 公司,创始人延续至 2026 年
2022-02Series D 融资完成融资$110M;累计融资 $250MTCV、Alpha Intelligence Capital、CDIB Capital 等投资方为 Aidoc 提供资本,使其从影像 AI 扩向更广泛医院平台野心
2022-03脑动脉瘤 510(k) K213721 获批监管FDA 510(k) 获批U.S. FDA显示早期神经血管监管进展
2023-11CAC 定量 510(k) K231631 获批监管FDA 510(k) 获批U.S. FDA扩展到定量和预防心脏病学工作流
2025-02CARE1 肋骨骨折许可公布监管基础模型设备许可Aidoc / Mercy 客户引言标志 Aidoc 从窄 AI 转向基础模型商业化
2025-07CARE 增长融资完成融资$150M 加 $40M 循环信贷额度;累计融资 $370MGeneral Catalyst、Square Peg、NVentures、战略医疗系统为 CARE 扩展和开放平台扩张提供资金
2025-07Globes 访谈拒绝披露估值反向估值未披露Aidoc 领导层 / Globes阻止确认当前独角兽状态
2025-12Asklepios 部署覆盖 28 家医院规模德国集团范围部署完成Asklepios Group、Aidoc证明跨国企业部署已超出试点范围
2026-01Hartford 合作三周内上线合作17 个算法在数百万次检查中上线Hartford HealthCare、Aidoc显示快速企业实施和广泛跨科室覆盖
2026-03综合身体 CT 分诊许可公布监管一个工作流中包含 11 个新适应证 + 3 个既有适应证Aidoc / U.S. FDA强化 CARE 围绕多适应证基础模型分诊的叙事
2026-04Series E 完成融资$150M;累计融资超过 $500MGoldman Sachs Alternatives、General Catalyst、SoftBank Vision Fund 2、NVentures 等投资方提供最新披露资本和投资人组合
2026-05真实世界 PE 研究摘要凸显局限反向算法在 97.8% 扫描中与放射科医生一致,但漏掉 15% 已确认 PENorthwell Health / dotmed 摘要强化结论:尽管分诊表现强,人类监督仍不可或缺

该时间线只使用抓取来源直接支持的里程碑。它同时纳入利好和反向事件,因为本章是后续章节的事实记录。

[CO001, CO015, CO017, CO019, CO027, CO028]
FO001: 公司里程碑时间线

从创立到基准运行日的关键公司、监管、客户和反向里程碑。

来源未给出具体日期时使用月份级日期。时间线纳入估值不透明和真实世界表现局限,因为二者对尽调有实质影响。

[CO001, CO015, CO017, CO019, CO027, CO028]
FO003: 快照 KPI

Aidoc 在基准运行日的当前规模、融资和证据状态指标。

该图有意混合完全佐证的事实和带标记的公司口径指标。病例、患者和医疗系统数量不可互换,不应轻率合并。

[CO021, CO022, CO023, CO024, CO025, CO037]

1.5 反向信号与未解决证据缺口

本次获取到的最强负面证据不是诉讼或警告信,而是一组技术和市场约束,它们限制了外界应如何激进解读公司的主张。Aidoc 自己的模型卡比营销页更坦率地说明了哪些内容仍不可观测:重要的人口统计属性不在 DICOM 中,因此偏差监测必须依赖代理变量和上市后性能跟踪,而不是完整的直接测量。外部证据也强化了谨慎必要性。2026 年 5 月一项真实世界肺栓塞研究摘要称,Aidoc 算法在 32,501 次扫描中与放射科医生的一致率为 97.8%,但也漏掉了 15% 的确诊 PE 病例,并在多数裁决分歧中落败,说明仍需人类监督。另一个问题来自通用放射 AI 的报销文献:当前支付框架仍很难适配这类产品,因此 Aidoc 的 ROI 叙事可能更多依赖避免错误、提升吞吐和下游照护管理,而不是直接支付代码。对公司层面尽调最重要的是,四个问题仍未解决:当前估值 / 独角兽状态、稳定的当前 FDA 批准数量、可支撑的当前员工数,以及有来源支撑的 170+ 模型数量。因此,本报告把 Aidoc 视为一家资本充足、规模较高且具备重要监管证明点的私营公司,但不把它视为一家估值和覆盖指标可无条件接受、披露已经完整的企业。[CO036, CO039, CO040, CO043, CO044, CO045]

Chapter 02

02市场分析

2.1 市场边界与纳入支出

分析 Aidoc 时,应先把它放在放射 AI 工作流市场内,再延伸到相邻的照护协同和跨专科编排层。官方放射科页面明确写到,公司价值从影像工作流中的分诊、优先级排序、定量发现和随访触发开始。照护协同页面随后把视角从放射科医生生产力扩展到多学科决策支持,尤其是肺栓塞等急症场景,时间、沟通和随访管理与影像解读同样重要。这一边界排除了通用 EHR 厂商、影像硬件 OEM 收入、计费软件以及大多数面向消费者的数字健康工具,即便这些系统仍是集成依赖。它也意味着,宽泛的“医疗 AI” TAM 说法会高估相关性,除非能被转换成放射科或急症工作流预算。MarketsandMarkets 和 Emergen Research 仍有用,因为它们确认医院是最大的放射 AI 买方,且以 CT 为中心、工作流属性强的用例主导当前支出。但本章把这些报告视为品类天花板,而不是 Aidoc 可获取市场的直接代理。关键分析动作,是把 Aidoc 的真实市场定义为依附于影像、分诊和下游照护触发的企业级工作流软件,而不是全部诊断 AI,更不是全部医疗 IT。[CM001, CM002, CM003, CM010, CM011, CM016]

市场定义表
细分 / 类别纳入支出排除支出买方 / 付款方相关性
放射科工作流 AI影像工作流内的分诊、优先级排序、定量、随访激活影像硬件收入、扫描仪更换、通用 PACS 许可证放射科、CMIO/CIO、企业运营Aidoc 的核心切入点和当前最清晰预算线
跨专科照护协同多学科激活、患者管理、紧急随访编排通用患者门户、无影像触发的呼叫中心 SaaS医疗系统和服务线领导层重要扩张层,把预算归属从放射科拓宽出去
企业临床 AI 操作层模型编排、治理、监控、分析和平台部署无编排的单算法独立点工具正在标准化 AI 治理的大型医疗系统对 Aidoc 的 aiOS 叙事和平台差异化至关重要
相邻肿瘤 / 心血管工作流与癌症负担、CAC、血管和心脏病学扩张相关的用例药物发现、治疗、整体肿瘤 EHR 系统专科科室和企业影像项目合理相邻领域,但还不是最清晰的公开收入核心
排除的一般医疗 IT账单、ERP、通用 EHR、广义临床文档床旁影像分诊和照护团队激活广义医院 IT 预算即便技术上集成,也在市场边界之外
现状替代方案:人工工作流放射科医生阅片队列、人工通知链、照护团队传呼和随访自主诊断主张承担人力和延迟成本的医院Aidoc 试图压缩而非完全替代的既有工作流

边界划在连接影像和下游行动的软件及工作流层。广义“医疗 AI”和硬件类别被有意排除,除非它们明确与 Aidoc 的部署经济性重叠。

[CM001, CM002, CM003, CM010, CM032, CM040]

2.2 受证据约束的市场规模视角

公开证据足以建立市场边界,却不足以把 Aidoc 的市场压成一个精确的 TAM/SAM/SOM 数字。自上而下视角先看美国医疗服务规模:CMS 报告 2024 年全国医疗支出为 $5.3 trillion,其中医院支出 $1.63 trillion,医生和临床服务 $1.11 trillion。医院数量视角再加上 AHA 的 6,100 家美国医院和超过 900,000 张配备人员的床位。品类视角随后缩窄到放射 AI:MarketsandMarkets 预计全球市场将从 2025 年 $0.76 billion 增至 2030 年 $2.27 billion,CAGR 为 24.5%;Emergen 则称医院在 2024 年约占全球需求的 55%。最后,装机基础视角看 Aidoc 本身:2025 年 CARE 融资稿中的 150+ 个医疗系统,2026 年 Series E 材料中的近 2,000 家医院,以及 Hartford 和 Asklepios 的已验证企业级部署。这些数字说明市场真实存在,也说明 Aidoc 已在具备企业级购买能力的买方子集中占据有意义份额。它们没有揭示的是每家医院、每个模块或每项检查的价格。正因为缺少定价数据,本报告使用低 / 基准 / 高代理区间,而不是假装知道每家医院或每条服务线的干净收入池。[CM006, CM007, CM017, CM018, CM037, CM038]

TAM / SAM / SOM 或规模测算视角表
视角地理 / 范围数值方法 / 置信度局限
总医疗支出上限美国医疗系统2024 年 US$5.3T高置信度;直接 CMS 支出总额过宽,无法直接映射到放射科 AI 收入
医院支出上限美国医院2024 年 US$1.6347T中置信度;直接 CMS 类别支出包含劳动力、设施和远超 AI 软件的服务线
医院基础美国服务提供机构6,100 家医院 / 907,216 张配备人员床位高置信度;直接 AHA 统计机构数量不揭示软件预算或 IT 准备度
全球放射科 AI 市场全球类别2025 年 US$0.76B 至 2030 年 US$2.27B中置信度;分析师市场报告类别范围仍包括 Aidoc 核心产品组合之外的供应商和用例
放射科 AI 中的医院份额全球类别拆分2024 年市场的 ~55%中置信度;分析师细分份额细分份额不等于 Aidoc 专属 SAM
Aidoc 披露的装机基础代理指标公司覆盖范围2025 年 150+ 家医疗系统;2026 年近 2,000 家医院中置信度;公司披露未披露价格、模块组合,也未披露如何把医院数量换算为收入
证据受限的美国 SAM 代理测算医院系统内的企业级放射科 + 护理协调软件低:US$4B / 中:US$8B / 高:US$16B置信度低;代理区间来自医院支出的极小比例,而非披露价格必须这样处理,因为未获取到 Aidoc 公开定价或品类层面的合同基准

本章不用一个笼统的 TAM,而是从多个视角拆分。最后一行是证据受限的代理区间,不是已披露的市场规模; 应把它当成预算启发,而不是干净的外部估算。

[CM006, CM007, CM017, CM018, CM037, CM038]
FM001: 从边界到规模金字塔

从广义美国医疗支出到 Aidoc 受证据约束的企业工作流机会的嵌套视角。

金字塔混合了已报告市场和支出上限,以及一个明确的代理层。它旨在说明,广义医疗 TAM 会夸大相关性,而纯放射科 AI 收入池可能低估工作流机会。

[CM001, CM006, CM007, CM018, CM037, CM040]
FM002: 市场估计区间

Aidoc 投资逻辑中最有用市场量的低位、中点和高位估计。

最后一行是示意而非字面值,因为 Aidoc 披露的医院数量是全球口径,而非仅限美国。该图意在展示规模敏感性,不声称精确市场份额。

[CM006, CM018, CM037, CM038, CM039, CM047]

2.3 买方地图与采用路径

最强的公开部署证据表明,Aidoc 面向大型服务提供机构销售,而不是依赖直接支付方报销或消费者渠道。Hartford HealthCare 把交易描述为围绕跨科室协作、AI 治理,以及放射科、心脏科、血管科、神经科和急诊科的照护延误减少。Asklepios 则把部署放在放射科人手短缺、小型站点支持和 24/7 急诊覆盖的语境下。由此推导的买方地图中,放射科负责人、CMIO、CIO 和医疗系统运营团队都很重要;预算归属会随着产品被作为科室级放射工具还是企业级平台采购而变化。Aidoc 自身营销进一步强化这一点:公司强调一次集成、多算法编排,以及把放射信号延伸到患者管理。这个组合使采用不再只是赢下一场单算法比拼,而是要证明平台能降低影像工作流各环节的摩擦。公开部署证据也显示,采用路径遵循标准企业软件模式:问题识别、集成设计、试点验证,再到全系统铺开。Aidoc 已披露的装机基础足够大,商业问题不再是产品能否走出试点,而是买方预算和治理结构能否支持更广泛的企业级标准化。[CM004, CM005, CM019, CM032, CM033, CM034]

细分市场 / 买方地图
细分市场买方使用者支付方 / 预算所有者工作流采用触发因素
学术医疗中心和大型整合医疗系统放射科主任、CMIO、CIO、创新负责人放射科医生、中风团队、急诊医生、服务线协调员企业运营或临床转型预算高量急性影像 + 下游协调减少积压、人手短缺压力、需要集中治理多种 AI 工具
社区医院网络放射科医疗主任、运营负责人普通放射科医生、急诊人员、转诊中心人员医院运营预算夜间 / 周末覆盖和时效性分诊现场专科团队更小,仍需守住质量
美国以外的私立医院集团集团数字化负责人和放射科负责人多院区放射科医生和急诊临床医生集团数字化预算或现代化项目跨多家医院集中部署多个院区需要统一标准平台
服务线护理路径肺栓塞、血管、心脏、神经条线负责人护理协调员和多学科团队有企业 IT 支持的服务线预算AI 发现出现后拉起正确团队希望把影像发现转成快速干预和随访
战略客户投资方参与融资轮的医疗系统临床和管理层推动者战略投资与部署预算组合使用平台,同时影响产品路线图希望抢先使用并获得企业共创影响力

公开部署案例最能证明大型医疗系统和医院集团。买方地图中的支付方一侧较弱,因为已获取证据更关注服务提供方采购 和提供方 ROI,而不是单独的保险方报销。

[CM032, CM033, CM034, CM035, CM036, CM040]
FM003: 按买方原型划分的部署路径

关系图展示买方原型、工作流和上线模式如何在 Aidoc 最可见的提供方细分中呈现差异。

公开案例由提供方主导且偏企业级,因此本轮对支付方特定采购路径的证据仍不足。

[CM004, CM005, CM019, CM032, CM033, CM034]
FM004: 买方采用漏斗

Aidoc 类部署的示意性企业采用路径,从识别工作流痛点到全系统标准化。

阶段百分比仅为示意,用来说明转化摩擦,并非经审计的赢单率。瓶颈更多在报销、集成和治理审批,而不是买方认知不足。

[CM019, CM020, CM032, CM033, CM035, CM036]

2.4 驱动因素、约束与报销现实

证据库中反复出现三类需求驱动。第一,劳动力压力是真实的:AAMC 仍预测医生短缺会延续到 2036 年,Neiman HPI 和 ACR 也称放射科医生短缺持续存在,且 2020 年后流失明显加速。第二,影像需求继续上升:Neiman 的需求研究预计到 2055 年影像利用率还会增加 16.9% 至 26.9%,老年人对影像的使用占比更高。第三,品类增长仍强:MarketsandMarkets 仍预计全球放射 AI 市场到 2030 年的 CAGR 为 24.5%。这些顺风之外,最清晰的约束是报销。同行评审报销论文和 Radiology Business 文章从不同角度提出同一点:多数影像 AI 工具没有单独付费的 CPT 代码,只有极少数较新的影像 AI 程序达到 Category I 状态。因此,ROI 必须来自吞吐、避免漏诊、减少重复检查或改善下游照护管理,而不是一条清晰的一对一支付流。监管负担是第二个约束。FDA 的 action-plan 和 PCCP 材料显示,自适应医疗 AI 仍需要严格的变更控制、监测和文档。第三个约束是集成摩擦:医院需要能适配旧 PACS/RIS/EHR 环境的系统,而不只是独立算法准确率高。[CM012, CM013, CM020, CM021, CM022, CM023]

增长驱动与约束表
驱动因素 / 约束方向时间影响尽调问题
放射科医生持续短缺利好需求结构性 / 多年期支撑 AI 分诊和工作流自动化采购量化目标买方本地短缺强度,不要只依赖全国平均值
影像使用量增长利好需求结构性 / 长期每名放射科医生处理更多扫描,提升优先级排序和编排的价值验证哪些模态和服务线造成最尖锐的积压痛点
医院在放射科 AI 采购中占主导利好需求当前契合 Aidoc 的企业级医疗系统销售模式测试较小医院缺少规模经济时,是否仍能支撑企业级定价
大多数影像 AI 的付费 CPT 覆盖稀疏负面 / 约束当前拖慢品类采用,迫使 ROI 来自运营节省梳理 Aidoc 具体急性适应症的直接报销状态
自适应 AI 治理和 PCCP 负担负面 / 约束当前 / 周期性抬高供应商生命周期成本,也增加买方验证工作核查哪些 Aidoc 产品已有成熟的变更控制计划
传统 PACS/EHR 集成摩擦负面 / 约束当前带来采购慢、IT 负担和试点疲劳按买方类型和已安装 IT 栈确认实施工作量
不一致病例仍需人工监督负面 / 约束当前限制自主性叙事,并保留放射科医生在环的工作流设计外推 ROI 前,先审查站点级表现和裁决数据
走向全系统临床 AI 平台利好需求当前 / 近期利好在编排、治理和客户成功上有深度的供应商验证客户是否标准化采用 aiOS,还是仍并行购买点状方案

主要商业瓶颈仍是报销,其次是集成成本和当前脱离人工监督使用的边界。多个利好因素是结构性的, 但近期转化仍取决于医疗系统的预算行为。

[CM012, CM013, CM020, CM021, CM027, CM028]

2.5 反向证据及其对采用节奏的含义

这个市场中最有用的反向证据是实际问题,而不是耸动故事。Aidoc 的真实世界肺栓塞研究摘要就是一个好例子:97.8% 的一致率听起来很强,但算法仍漏掉了 15% 的确诊 PE 病例,并在多数裁决分歧中落败,医院因此无法负责任地把 AI 当作放射科医生替代品。这与其说是产品失败,不如说是在提醒市场:价值主张是分诊和工作流支持,不是自主阅片。报销证据也指向同一方向。如果多数急症放射 AI 用例无法获得单独付费的 CPT 代码,买方就会继续靠运营节省和临床风险降低来证明采购合理性,而不是靠直接科目报销。这会拖慢采用,因为企业软件预算、IT 资源和临床治理委员会都必须先相信效率与安全叙事,产品才能规模化。Aidoc 当前装机基础说明,这道门槛对成熟医疗系统可以跨过;但它也意味着未来增长取决于能否赢下中央工作流预算、证明集成 ROI,并展示基础模型治理可以被长期信任。换句话说,这个品类已经跨过可信度门槛,但尚未跨过无摩擦预算门槛。[CM025, CM026, CM027, CM028, CM029, CM030]

Chapter 03

03竞争格局

3.1 竞争版图与现状替代方案

Aidoc 并不处在一个边界整齐的单一品类中。它的产品栈从受监管的放射科分诊开始,延伸到 aiOS 编排,再进入照护协同和下游随访。因此,真正的对比对象包括 Viz.ai 这样的直接临床 AI 同行,Enlitic 和 Nanox.AI 这样基础设施优先的影像厂商,PathAI 和 Paige 这样相邻病理基础模型玩家,以及已经掌握 PACS、归档、路由和安全审查的既有影像 IT 厂商。独立行业报道也强化了这一点:医院越来越为工作流集成和企业级价值买单,而不是为孤立的点算法买单。 现状替代方案仍很强。ACR 的 ARCH-AI 计划把影像 AI 部署界定为持续的治理和质量保证纪律,而不只是一笔软件采购。Radiology Business 也认为,许多医疗系统如果看不到 AI 在放射科之外改善企业运营,仍难以证明采购合理性。实际采购中,这意味着既有 PACS、工作列表和院内治理仍是 Aidoc 的真实替代品,尤其当医院可以在现有基础设施上渐进添加 AI,而不是替换其工作流层时。 [CP001, CP002, CP003, CP004, CP031, CP032]

竞品画像表
竞品组类别工作流层规模 / 商业化信号相对 Aidoc 的核心优势相对 Aidoc 的主要限制
Viz.ai直接急性护理临床 AI叠加在影像触发工作流上的护理协调平台50+ 个 FDA 获批算法;Series D 轮时覆盖 1,000+ 家医院;后续引用 1,400+ 家医院 / 2.2 亿人群最接近的同类组合:影像检测 + 下游护理团队调动更偏疾病项目中心;在开放式多服务线编排上不如 Aidoc 讲得明确
Enlitic + Annalise影像数据基础设施 + 广泛发现支持面向放射科的标准化、报告和分诊工作流工具Enlitic 面向放射科医生 / PACS 管理员;Annalise 入选 64 家 NHS 信托机构,每年处理 280 万张胸部 X 光在影像数据质量和多发现解读支持上覆盖面强影像工作流之外的企业级护理协调,公开证据较少
Nanox.AI / Zebra影像网络 + AI/软件混合模式AI 和软件挂在更广的影像及远程放射栈上上市公司;2025 年 Q4 AI/软件收入仍只有 $0.5M可把 AI 与硬件、影像网络和远程放射动作结合商业化仍早期,验证程度远低于医院工作流营销所暗示
PathAI数字病理平台云端病理工作流和 AI 应用中枢AISight 平台;2026 年被 Roche 收购,前期付款 $750M,另含里程碑付款病理工作流和生物制药定位强不是放射科或急性护理协调平台
Paige计算病理 / 基础模型进入者病理医生助手和生物标志物工作流工具近 700 万张数字化切片;被 Tempus 以 $81.25M 收购数据资产强,病理基础模型叙事也强与模态相邻而非放射科优先;医院工作流覆盖面窄得多
Nuance / Microsoft工作流分发和 AI 市场层Precision Imaging Network 建在既有影像和云工作流关系上单一合同 / BAA / MSA 提供部署捷径;合作伙伴生态触达保险计划和雇主简化采购,拥有既有工作流关系目前更像分发层,而不是端到端临床 AI 运营模式
Sectra / Fujifilm / AGFA / GE / Intelerad 等影像 IT 既有厂商PACS 和企业影像既有厂商既有 PACS、云归档、工作流编排和嵌入式 AI 界面Sectra 和 Fujifilm 推广 AI 市场 / 编排器;GE 对 Intelerad 估值 $2.3B,且 90% 为经常性收入装机基础、捆绑采购,以及跨放射科及更广场景的工作流控制单个算法可能弱于 Aidoc 第一方临床栈,或更依赖合作伙伴
现状 / 院内治理既有 PACS + 人工升级 + 选择性 AI 插件医院自有工作清单、路由、QA 和治理委员会ROI 或治理证明薄弱时,仍是默认选择切换痛感最低,本地控制最大碎片化继续存在,也可能拖慢企业级随访工作流

公开信号强调工作流位置、装机基础杠杆和战略方向。实际定价和许多私营公司收入仍未披露。

[CP007, CP009, CP011, CP014, CP016, CP018]
FP001: 竞争定位图

Aidoc 在临床覆盖广度与企业工作流控制的组合上得分最高;传统厂商主导工作流控制,专科厂商则主导更窄的模态。

x 轴是有证据支撑的临床广度,y 轴是有证据支撑的工作流控制;评分来自保留公开来源基础上的有序分析判断,不是实测产品基准。

[CP002, CP004, CP008, CP011, CP017, CP019]

3.2 直接临床 AI 同行

Viz.ai 是最清晰的同类竞争对手,因为它把影像触发检测与下游照护团队动员结合起来,并且已经进入大型医疗系统部署。它营销 50+ 个 FDA 批准算法,把自己定位为照护协同平台,而不是单一用途工具,也有公开证据显示其完成独角兽规模融资。这使 Viz.ai 成为最直接的标尺,用来判断在更多厂商增加算法广度后,Aidoc 能否继续保持急症照护协同差异化。 Enlitic 和 Annalise/Harrison.ai 的竞争方式不同。它们的信息更强调影像数据标准化、工作流效率和广泛发现覆盖,而不是全系统照护协同。Nanox.AI/Zebra 又处在另一条相邻赛道:它把 AI 和软件与更宽的影像网络、硬件议程结合起来。这些同行很重要,因为即使它们没有完全复制 Aidoc 跨专科运行层的叙事,也可能在模态广度、算法数量或医院工作流嵌入上挤压 Aidoc。 [CP005, CP006, CP008, CP009, CP010, CP011]

功能 / 能力矩阵
采购标准AidocViz.aiEnlitic / AnnaliseNanox.AIPathAI / PaigeNuance / MicrosoftPACS 既有厂商
急性影像分诊None
跨专科护理协调None低-中
第三方模型托管 / 生态系统
病理覆盖NoneNoneNoneNoneNone
嵌入式 PACS / 影像 IT 控制
治理 / QA 工具侧重
开放式企业 AI 运营层叙事

评级是基于留存官方与独立来源的证据支持的序数判断;它们总结买方契合度, 而非每个用例中的临床优越性。

[CP002, CP003, CP004, CP008, CP010, CP011]
FP002: 功能广度 / 能力图

买方任务矩阵展示 Aidoc、直接竞品和影像传统厂商在医院 AI 采购标准上的强项分布。

高 / 中 / 低 / 无均为有证据支撑的有序判断,概括各竞争者当前最容易赢单的位置,不是完整功能清单。

[CP003, CP006, CP013, CP017, CP021, CP023]

3.3 病理进入者与工作流既有厂商

PathAI 和 Paige 不是放射优先,但它们仍然重要,因为它们争夺同一笔企业 AI 预算和治理注意力。两家公司都主打病理基础模型、工作流工具和大型数据资产;也都在 2025-2026 年成为收购目标,凸显工作流加模型这一层已经具备战略价值。它们与 Aidoc 的关系,不在于一套病理产品能替代急症放射分诊,而在于医院高管越来越把 AI 作为跨服务线的运行层决策来评估。 更持久的分发威胁来自既有影像 IT 厂商。Nuance/Microsoft、Sectra、Fujifilm、AGFA、GE 和 Intelerad 都把 AI 部署营销为既有影像基础设施的延伸,经常强调单一合同上线、云 PACS 或嵌入式编排,而不是单个算法。一旦医院相信工作流层应承载多个第三方模型,既有厂商就会凭装机基础、采购熟悉度和集成控制权获得杠杆;而 Aidoc 仍必须逐个账户赢下来。 [CP017, CP018, CP019, CP020, CP021, CP022]

定价 / 打包比较
竞品组主要买方公开打包信号部署动作合同 / 续约影响证据缺口
Aidoc医疗系统和放射科主导的企业级买方私人报价市场上架;平台 + 护理协调叙事深度工作流集成和企业级部署一旦跨模态集成,可能形成多年经常性关系无公开标价或实际成交价格
Viz.ai医疗系统、中风 / 心脏项目、生命科学以独角兽级私募估值融资,支撑平台扩张疾病项目式部署,正扩展到更多病种若临床团队围绕项目化路径标准化,粘性强公开定价仍不透明
Enlitic / Annalise放射科、影像网络、医院 IT工作流和数据标准化定位;Annalise 赢得 NHS 采购在科室或网络层面接入报告和分诊数据规范化和报告流程嵌入后,粘性强实际合同结构公开披露有限
PathAI / Paige实验室、病理集团、生物制药、癌症中心企业级病理平台和 AI 授权由数据集和工作流牵引,部署进病理栈若纳入诊断和生物标志物工作流,粘性强与放射科合同相邻,而非直接替代
Nuance / Microsoft / PACS 既有厂商CIO、影像 IT、企业影像负责人单一合同 / BAA / MSA,或 PACS 捆绑的 AI 界面在既有影像和云关系中加入 AI既有基础设施已深度扎根时,续约杠杆最高难把 AI 价格从更广的影像 IT 支出中剥离
现状 / 选择性插件本地放射科领导层和治理委员会增量插件采购、试点或人工工作流调整缓慢按用例逐一部署承诺度最低,企业级标准化也最弱节省额往往难以持续一致地衡量

公开披露更清楚地显示打包逻辑和采购姿态,而非实际价格。医院常先买工作流层, 然后才有能力评估算法标价。

[CP007, CP009, CP011, CP014, CP017, CP019]

3.4 商业化、切换成本与护城河耐久性

Aidoc 仍拥有一组真实竞争资产。官方材料显示,其能力覆盖放射科分诊、企业级编排和照护协同;2025 年融资材料还显示,客户已经在 aiOS 上运行第三方模型。这支撑了一个判断:Aidoc 正从单算法厂商走向操作系统。战略上这很重要,因为医院越来越希望减少供应商数量,并获得更一致的治理。 但护城河并非绝对。多供应商编排正在成为基本要求,而不是独特能力。Fujifilm 营销开放编排和 50 多个已验证算法,Nuance/Microsoft 借既有工作流合同销售更轻松的部署,Sectra 加 GE/Intelerad 则能把 AI 与更宽的影像基础设施打包。独立报道也清楚指出,报销和直接 ROI 对许多影像 AI 仍然狭窄。因此,当医疗系统重视集成随访、偶发发现和企业 AI 治理时,Aidoc 可以赢;但在采购由既有工作流控制权主导,或医院把 AI 视为 PACS 内功能而不是新运行层的场景中,Aidoc 很脆弱。 [CP006, CP007, CP021, CP023, CP026, CP029]

护城河耐久性 / 竞争风险登记表
护城河主张重要性威胁严重度尽调问题
受监管分诊 + 编排 + 护理协调让 Aidoc 的叙事比单算法对手更宽Viz.ai 和既有厂商正向护理协调 + 工作流叙事靠拢询问收入中有多少来自编排 / 护理协调,多少来自第一方算法
开放生态和第三方模型托管帮助医院用一个运营层压缩供应商蔓延Fujifilm、Nuance/Microsoft 和 PACS 厂商也在推广开放式 AI 托管验证 aiOS 在启用速度和治理上是否显著优于既有厂商编排
深度工作流集成嵌入式 PACS / EHR / 报告连接抬高切换成本既有影像厂商已经控制大部分管线索取相对既有 PACS 厂商的流失率、替换胜出和部署时长数据
部署规模和 FDA 历史传递信任和实施成熟度信号现在更多供应商声称拥有广泛获批和大型装机基础比较活跃使用率和续约,而不只是客户 logo 或获批数量
来自偶发发现和随访的企业级 ROI报销仍窄时,企业级价值就是核心采购论点独立证据显示,除非在全系统层面衡量,ROI 仍常常站不住要求提供经审计的客户 ROI 研究,覆盖范围需超出放射科周转时间
病理和多模态邻近性可能把 Aidoc 的预算相关性扩展到放射科之外PathAI、Paige 和其他多模态进入者可能先拿到 AI 战略预算澄清当前偏放射科优势之外的路线图可信度

本登记表关注护城河耐久性,而非产品质量。核心问题是,当既有厂商和相邻专科玩家扩张时, Aidoc 能否继续控制企业 AI 运营层。

[CP004, CP006, CP021, CP023, CP031, CP032]
FP003: 护城河 / 就绪度 KPI

浓缩呈现最直接影响 Aidoc 2026 年竞争就绪度的公开信号。

[CP006, CP021, CP029, CP030, CP033, CP042]

3.5 展示

Chapter 04

04财务情况

4.1 收入模式和定价信号指向企业软件,而不是透明交易定价

从各个角度看,Aidoc 的公开商业姿态都由企业销售牵引。产品页面描述了它与 PACS、EHR、排班、报告和照护协同工作流的深度集成;AWS Marketplace 页面则称该产品只通过私下报价提供。这是协商式企业合同的典型特征,而不是自助软件或标准化按检查结账流程。Aidoc 的变现面看起来也比单一放射科分诊模块更宽:aiOS 治理多个模型,放射科产品嵌在既有影像工作流中,照护协同层则把价值延伸到患者随访和跨专科触发。 缺失的内容同样重要。官方页面不发布标价、按检查费率或分部收入结构。本次找到的唯一具体公开定价信号,是一个可靠性较低的第三方评论页面,估计首年成本为 $50,000 到 $150,000 以上,并称每次安装起价约 $50,000。作为包装代理,这个信号有方向性价值,但还不足以支撑投资测算。可辩护的结论是,Aidoc 的行为像带实施工作和工作流粘性的企业 SaaS / 站点许可软件,而实际成交价格和收入确认仍未披露。 [CI001, CI002, CI003, CI004, CI005, CI006]

收入流表
收入流机制单位当前公开证据收入质量判断尽调问题
企业级放射科 AI 部署授权临床 AI 嵌入放射科工作流医疗系统 / 医院合同官方页面强调深度 PACS 和 EHR 集成,以及私人报价销售一旦嵌入,可能经常性强且粘性高按账户规模索取合同期限、部署费和续约结构
aiOS 编排平台运行、治理并监控多个模型的操作层平台 / 站点许可式合同aiOS 被定位为企业平台,而不只是某个模块里的功能相比单一用例的算法销售,平台收入质量可能更高索取平台与算法收入拆分,以及多年期合同附加率
照护协调与患者管理调动随访和多学科团队的工作流工具跨专科工作流合同或平台附加包官方照护协调材料把价值落在下游行动和随访上如果医院买的是企业价值而非单纯分诊,钱包份额可能加深索取附加率、席位模型和结果挂钩定价敞口
第三方模型托管 / 生态系统aiOS 除 Aidoc 自有算法外,还托管外部模型平台费,或从企业编排价值中变现Aidoc 称 69% 的客户已经在 aiOS 上运行非 Aidoc 模型提升粘性,也可能在无需为每个用例投入自有研发的情况下提高单账户收入索取外部模型托管定价,以及收入属于软件还是代收代付
实施 / 集成服务将工作流接入 PACS、EHR、报告和治理栈一次性或分阶段实施服务深度工作流集成是公开定位的核心先落地再扩张离不开它,但如果服务占比过高,可能稀释毛利率索取实施成本回收情况,以及服务与软件的毛利率结构

公开证据能支持收入模式的大致形态,但不能支持各分部占比、收入确认或实际定价。

[CI001, CI002, CI003, CI004, CI007, CI008]
定价 / 变现表
触点公开定价信号含义可信度缺口
Aidoc 官方产品页面未公布标价买方需要走企业销售流程合同按医院、站点、影像模态还是整个网络计价
AWS Marketplace 列表仅私下报价企业级打包谈判,而不是透明的点击购买定价实际报价区间、最低承诺额和多年期条款惯例
ItQlick 评测网站首年估算 $50k-$150k+;声称每次安装 $50k很粗的外部代理指标,但指向可观的前期企业支出方法论、真实客户报价,以及估算反映的是 Aidoc 还是品类均值
照护协调 / 工作流 ROI 信息未披露定价;价值围绕医院效率和随访来表达变现可能取决于企业价值主张,而不是按扫描计费是否有合同包含收益分享、绩效保证或服务线增购经济性
aiOS 上的生态托管未公布第三方模型托管价目表平台经济性可见,但抽成率不可见外部模型的收入分成、平台费和毛利结构

官方证据显示销售通过企业级谈判完成。第三方定价页面只能作为较弱的方向性代理指标,不应视作管理层指引。

[CI003, CI004, CI005, CI006, CI023, CI043]
FI001: 收入模型桥

展示 Aidoc 的企业部署模式如何把临床 AI 能力转化为经常性平台价值。

该图反映的是公开商业化逻辑,而非披露的合同会计处理;它说明收入机制,不是经审计的报表项目。

[CI001, CI002, CI003, CI007, CI008, CI009]

4.2 资本历史可见,但估值支撑仍不完整且内部不一致

相比已披露的经营指标,Aidoc 最新融资记录要新得多。公司在 2023 年称,一轮 $110 million Series D 使总融资达到 $250 million。2025 年,一份 PRNewswire 新闻稿描述了 $150 million 融资加 $40 million 循环信贷额度,并称总融资为 $370 million。随后在 2026 年 4 月,Aidoc 和 Goldman Sachs 称一轮 $150 million Series E 使总融资超过 $500 million。单独看,每份公告对新增轮次都清楚。合在一起看,累计总额无法干净对上,因此建模已融资资本时,最安全的方法是逐轮相加,而不是相信标题式累计融资总额。 估值证据更薄。Calcalist 报道称,2025 年轮次的估值高于上一轮;Globes 则称公司不会披露收入或该轮估值。这样一来,2024 年后的记录支持动能,却不足以给出数字化公允价值判断。实际含义是,即便经过 2026 年 Series E,公开证据目前也不能确认 Aidoc 的精确估值或当前独角兽状态。 [CI010, CI011, CI012, CI013, CI014, CI015]

资本充足性表
项目公开数值 / 状态可信度含义尽调要求
2023 Series D 轮$110M;官方当时称累计融资为 $250M奠定了 2025 年之前可观的股权融资基础索取 2023 年以来的股权结构桥接
2025 增长融资$150M,另有 $40M 循环信贷;官方当时称累计融资为 $370M显示公司仍能获得资本,并引入类债务融资索取循环信贷完整条款、提款状态、契约限制和资金用途
2026 Series E 轮Goldman Sachs 领投 $150M;官方当时称累计融资超过 $500M显著降低近期融资压力,并为扩张提供资金索取交割后现金余额和现金跑道假设
估值披露2025 年报道称估值高于上一轮,但未披露具体数字动能存在,但公允价值支撑仍弱索取投后估值、清算优先权和任何参与型债务条款
现金、烧钱速度和现金跑道未公开披露无法真正评估资本充足性索取月度烧钱、净现金和下行情景现金跑道计划
资本强度触发点公司在上一轮融资后不到一年再次融资,同时全球扩张 CARE 和 aiOS说明激进扩张仍会消耗不少资本按资金用途索取模型开发支出、托管成本和招聘计划

分轮披露有用;但累计融资总额之间并不完全自洽,应直接向管理层交叉核验。

[CI010, CI011, CI012, CI013, CI014, CI015]
FI003: 财务估计区间

公开可观察的融资、定价与估值输入区间;这些是参考区间,不是承销后的 Aidoc 指标。

该图有意避免虚构 Aidoc 收入、ARR 或估值。它只展示有边界的公开参考,用来辅助尽调框架。

[CI011, CI013, CI034, CI035, CI036, CI037]

4.3 单位经济逻辑可信,但公开数字止步于客户价值证明和外部基准

Aidoc 的公开价值叙事可信。官方材料和投资人评论把产品与住院时长缩短、放射科效率提升、诊断延误减少,以及医疗系统可衡量财务回报联系起来。这很重要,因为影像 AI 单靠报销仍难以证明自身价值。独立市场报道称,只有少数影像 AI 应用具备有意义的报销,且除非收益覆盖更广的医院网络,否则 ROI 往往站不住。Aidoc 自己的销售话术与这一现实一致:产品被作为工作流基础设施和照护协同销售,而不只是一个可报销的影像分析小组件。 公开证据止步的地方,是厂商 P&L 内部。公司没有披露毛利率、CAC、回本周期、NRR 或现金转换。为了锚定经济逻辑,本次最好的公开基准是 GE 计划收购 Intelerad 的交易,它意味着成熟的工作流层影像软件可以产生约 90% 经常性收入和超过 30% adjusted EBITDA 利润率。Nanox 提供相反的警示样本:一家上市影像 AI 公司仍可能只有很小的软件收入、亏损和持续融资需求。Aidoc 可能位于这两个极端之间,但公开证据没有显示具体位置。 [CI019, CI020, CI021, CI022, CI023, CI026]

单位经济性表
指标公开数值 / 状态可信度为什么重要尽调要求
客户 ROI 证据公司声称可缩短住院时长、提升放射科效率,并带来可衡量的财务回报;但公开渠道没有统一的经审计 ROI 研究支撑需求质量和续约逻辑按部署队列索取经审计的前后对比经济性
报销依赖公开证据显示,影像 AI 采用仍更多依赖系统级效率,而不是 CPT 报销决定 Aidoc 卖的是预算节省,还是创造可计费收入按产品线索取报销敞口,以及 ROI 保证相关合同条款
切换成本与 PACS、EHR、报告和治理栈深度集成,意味着实施摩擦不小软件一旦粘住,可支撑更高倍数和更好的续约经济性索取流失、扩张和替换竞品赢单数据
平台附加 / 生态经济性据报道,69% 的客户在 aiOS 上运行非 Aidoc 模型平台托管可提升单客收入和战略相关性索取外部模型托管费和附加率队列
供应商毛利率路径未公开披露是判断软件质量和估值的核心索取软件与服务毛利率,以及托管 / 推理成本趋势
销售效率和回本周期未公开披露决定增长是资本高效,还是依赖融资按细分市场索取 CAC、销售周期、实施时间和回本周期

这张表把买方价值证据和供应商层面的经济性拆开。公开来源更擅长证明前者,后者证据更弱。

[CI008, CI009, CI019, CI020, CI021, CI022]
FI002: 单位经济桥

Aidoc 的公开经济性在客户价值证明上最强,在厂商利润率披露上最弱。

该桥图用公开价值信号和市场质疑来展示因果逻辑;它不能替代已披露的 CAC 或利润率数据。

[CI022, CI023, CI026, CI027, CI028, CI029]
FI004: 资本强度 / 现金流图

梳理最重要的公开财务信号、它们意味着什么,以及仍阻碍承销的证据缺口。

[CI014, CI028, CI029, CI038, CI039, CI040]

4.4 结论:需求信号强,投资测算文件弱

以当前证据看,Aidoc 的财务画像强于早期算法厂商,但弱于一家足以完整测算的后期软件公司。2025 年融资、2026 年 Series E 和广泛部署主张说明市场拉力有意义,也降低了短期财务困境风险。Series E 之后的新闻流仍保持活跃,Aidoc 强调新客户部署、欧洲产品扩张和一位全国知名 CMO 加入,而不是释放收缩信号。产品覆盖、集成深度和企业级包装都符合经常性、高切换成本软件特征。相邻 AI 诊断领域的公开可比交易——Viz.ai 的独角兽融资、Roche 收购 PathAI,以及 Tempus 收购 Paige——也显示战略买家和成长投资人愿意为工作流、数据和模型资产的组合付费。 但对财务买家最在意的每一项指标,文件仍不完整。收入、ARR、毛利率、烧钱速度、现金、债务、CAC、回本周期和 NRR 都未披露。公开定价不透明。公司自己的总融资数字在官方新闻稿之间相互冲突。2024 年后的记录也仍未公开确认数字化估值或当前独角兽状态。因此,正确的财务立场是:认可商业化质量和平台相关性,但在管理层打开账本之前,对利润率路径、估值支撑和资本充足性保持明确谨慎。 [CI014, CI015, CI030, CI031, CI034, CI035]

公开财务缺口表
缺失的私有指标对投资判断的影响最佳公开替代指标具体尽调路径
收入 / ARR无法衡量核心软件业务规模或增长质量只有部署数量和患者量声称索取月度经常性收入桥接,以及按产品族拆分的 ARR
毛利率无法判断软件质量或服务占比只有 GE / Intelerad 工作流软件基准索取软件与服务毛利率,以及托管成本趋势
现金余额和烧钱速度无法判断现金跑道或下一轮融资时点最新融资公告降低了困境风险,但不能替代现金数据索取最新资产负债表、烧钱速度和契约限制时间表
CAC / 回本周期 / 销售周期无法判断增长的资本效率企业集成深度意味着部署成本不低按细分市场索取漏斗转化、实施成本和回本周期
留存 / NRR / 扩张无法判断 aiOS 能否成为持久的操作层aiOS 上 69% 的外部模型使用率值得关注,但信息不完整索取 logo 留存、GRR、NRR 和外部模型附加率队列
精确估值无法凭 2024 年后的证据确认公允价值或当前独角兽身份公开信息只报道“高于上一轮”索取最新一轮条款清单或董事会批准的估值备忘录
收入确认政策无法判断收入中有多少是经常性软件,多少来自服务、实施或绩效挂钩工作官方信息显示,收入来自企业软件加工作流价值索取收入确认备忘录和客户合同样本

这些不是表面缺口;正是这些具体项目,让 Aidoc 仅凭公开证据还无法从“有潜力”推进到足以做投资判断。

[CI005, CI009, CI014, CI017, CI031, CI033]

4.5 展示

Chapter 05

05产品与技术

5.1 aiOS 平台架构与运行模式

Aidoc 的核心产品不只是一组点算法库,而是 aiOS,一个面向临床 AI 的企业级操作系统。aiOS 页面明确把平台描述为在医疗系统内运行、编排并治理临床 AI 的层,并用一次集成支撑多个工作流。实际含义是,医疗系统不需要把每个算法都部署成独立小组件。aiOS 会组织影像和元数据输入,判断哪些模型应该运行,在患者和检查语境中交付输出,并在系统层面提供治理、监控和结果跟踪。 这套架构以工作流为先。Aidoc 称 aiOS 使用文本数据、扫描元数据和像素分析来决定哪些扫描应接受哪些 AI 处理,然后把疑似发现推送到 PACS、EHR、移动端和照护团队工作流。Aidoc 也把 aiOS 定位成开放生态层,而不是封闭产品孤岛:外部模型托管是明确设计目标,多家独立 2025 年来源也称,多数客户已经在该平台上托管非 Aidoc 模型。这使 aiOS 更接近医院 IT 的 AI 控制平面,而不是单一用途放射科应用。 公开部署案例支持平台叙事。Hartford HealthCare 在三周内初步上线,Mercy 在 50 个设施中标准化 aiOS,Asklepios 完成了覆盖 28 家医院的部署。这些案例说明,产品已经从试点阶段的科室工具进入企业级实施模式。剩余注意点在于,Aidoc 官方网站仍营销“规模最大的 FDA 批准算法组合”,却没有发布当前官方数字化组合总数,因此描述广度时必须谨慎。 [CE001, CE002, CE003, CE004, CE005, CE006]

产品模块 / 资产矩阵
模块 / 资产主要用户状态 / 成熟度关键差异化尽调缺口
aiOS 企业平台CIO / 放射信息化 / 服务线负责人已在企业级规模上线通过一个操作层运行、编排和治理多个 AI 解决方案官网没有给出可对账的产品组合数量
CARE 基础模型临床 AI / 产品 / 放射科负责人已在上线产品中获 FDA 批准;仍在扩展多模态基础模型,扩大分诊覆盖,并复用开发能力独立跨站点基准语料仍有限
Neuro “Full Brain” 套件卒中、神经介入、放射科团队已商业化上线不受扫描协议和首要目的限制,运行多个神经算法各模块敏感性 / 特异性的公开披露不完整
VTE 解决方案放射科、PERT、血管科、ED 团队已商业化上线整合 PE/iPE 检测、DVT 警报、RV/LV 数据、聊天和随访工作流结果证据集中在部分标杆站点,而非整个产品组合
主动脉解决方案血管、心胸外科、放射科团队已商业化上线在一个工作流里打通急性夹层分诊、动脉瘤随访和照护激活未公开逐客户列明部署深度
照护协调多学科急性照护团队已商业化上线移动警报、影像查看、EHR 上下文和跨部门协作诊断 / 非诊断边界必须谨慎治理
患者管理随访协调员 / 专科诊所已在部分路径商业化上线基于文本识别动脉瘤和 IVC 滤器工作流里的随访对象纵向依从性结果的公开披露有限
开放生态 / 外部模型企业 AI 治理团队据公开表述已上线aiOS 在 Aidoc 应用之外托管第三方模型第三方模型的公开准入标准和质量门槛不详细

状态反映截至 2026-05-20 的公开证据。“已商业化上线”是指 Aidoc 公开营销该工作流并引用客户使用,不意味着每个客户站点部署深度相同。

[CE001, CE007, CE016, CE017, CE019, CE027]
技术 / 运营架构表
层 / 组件角色关键依赖主要风险
检查接入与元数据过滤将检查匹配到工作流标准,并准备输入DICOM 影像、模态元数据、检查路由规则标准配置错误可能压掉相关检查
编排层根据解剖部位和上下文选择每次扫描运行哪些模型aiOS 逻辑、元数据、像素和文本信号公开架构细节没有披露完整决策树
模型执行层运行 CARE 或其他任务专用算法GPU / 云 / 本地计算和模型注册表第三方模型质量治理未公开详细说明
通知 / 预览层发送优先级标记和预览图像PACS、工作列表、移动客户端预览输出不是诊断结论,不能替代全图审阅
照护协调层跨专科路由警报、聊天和团队激活EHR、移动端、基于角色的升级路径警报过多或配置不佳会造成工作流疲劳
患者管理层跟踪动脉瘤或 IVC 滤器等随访人群NLP / 报告抽取和排程连接纵向留存指标未广泛公开
治理 / 分析层监控采用率、性能和人工覆写验证、漂移检测、分析仪表盘公开领域的独立审计证据有限
安全 / 合规基础保护数据,并约束受监管用途AWS/Azure、NIST CSF、QMS、MDR/FDA 控制信任中心披露细节少于客户采购尽调通常要求

架构基于官方产品、安全、质量和 FDA 材料综合而成。Aidoc 尚未发布一张全面的参考架构图,能对齐每种客户部署变体。

[CE002, CE004, CE005, CE014, CE015, CE018]
FE001: 产品架构图

Aidoc 的架构把企业编排、临床模型、工作流交付和治理叠在既有医院影像与 EHR 基础设施之上。

分层综合了多个官方页面和 FDA 摘要,并非来自单一公开厂商参考架构图。

[CE001, CE004, CE007, CE015, CE021, CE022]
FE002: 客户工作流 / 运营流程

aiOS 把 AI 分诊和协调插入既有影像工作流,同时不替代标准照护读片。

流程反映 Aidoc 公开产品描述和 FDA 批准的运行边界;具体站点配置各不相同。

[CE002, CE004, CE014, CE018, CE020]

5.2 CARE 基础模型与解决方案广度

CARE(Clinical AI Reasoning Engine)是 Aidoc 从单病种算法走向可复用临床基础模型的下注。Aidoc 的基础模型页面称,CARE 使用真实世界多模态数据训练,不仅覆盖影像,也覆盖文本、EHR 字段、化验和生命体征。2026 年 1 月体部 CT 获批,是该模型不再停留在概念层面的最强公开证明点。Aidoc 官方新闻稿和 Diagnostic Imaging 的独立摘要都称,此次获批把 11 个新批准适应症与 3 个此前已批准适应症合并进一个工作流,形成 14 个适应症的体部 CT 分诊产品。 FDA 的 K252970 摘要提供了最具体的技术细节。获批设备是 BriefCase-Triage: CARE Multi-triage CT Body,这是一款面向胸部、腹部和骨盆增强及非增强 CT 的工作流分诊和通知产品。FDA 文件列出的 11 个新发现包括憩室炎、腹盆腔脓肿、阑尾炎、肠缺血 / 肠壁积气、梗阻性肾结石、小肠梗阻、大肠梗阻、脾损伤、肝损伤、肾损伤和骨盆骨折。Aidoc 新闻稿还报告称,在 FDA 审查研究中,11 个新适应症的平均敏感性为 97%,平均特异性为 98%。 在综合体部 CT 获批之外,Aidoc 产品页面也显示了既有产品栈的临床广度。神经页面捆绑了血管闭塞、CT 灌注、脑动脉瘤、颅内出血、颈椎骨折和椎体压缩性骨折工作流。VTE 页面从肺栓塞延伸到偶发 PE、DVT 和 IVC 滤器管理。主动脉页面把急性主动脉夹层分诊与动脉瘤随访结合起来。合在一起,这些页面支持一个围绕 PE、ICH、主动脉夹层、骨折和颈椎发现的真实工作流组合。仍不够清楚的是当前官方总组合数量:公开来源支持广度,但 Aidoc 自己的网站没有发布已调和的总数。 [CE007, CE008, CE009, CE010, CE011, CE012]

工作流 / 用例表
用户任务现有工作流问题Aidoc 解决方案可衡量收益限制
拥挤 ED / 门诊积压下的躯干 CT 分诊FIFO 阅片延误急性发现CARE Multi-triage CT Body 在一个工作流中标记共 14 项适应证报告称 11 项新增适应证的平均敏感性为 97%、平均特异性为 98%公开材料没有在一处列明此前已获批的全部 3 项适应证
卒中 / 神经急诊响应按协议设计的工具会漏掉偶发或非主要发现Full Brain 套件不受扫描目的限制,运行相关神经算法引用 Ochsner LSU 案例数据证明卒中工作流改善公开证据偏向精选案例
PE / VTE 升级处理人工升级拖慢治疗和随访PE、iPE、DVT 和 IVC 工作流,加上 PERT 激活和接入 EHR 的随访VTE 页面引用 Yale 和 Cedars 的结果案例结果具有站点特异性,且部分由公司呈现
急性主动脉照护时间敏感的夹层 / 动脉瘤工作流仍然碎片化分诊、移动警报、影像查看、聊天和动脉瘤患者管理Mount Sinai / Yale / HOAG 引述显示多学科使用未按站点公开拆分企业部署深度
企业 AI 治理独立工具造成集成蔓延aiOS 集中承载编排、验证、监控和分析Hartford、Mercy 和 Asklepios 展示多站点推广模式第三方模型准入规则未公开披露
未来报告自动化阅片积压和报告滞后仍是运营瓶颈CARE 路线图包含从像素到报告草稿的工作流路线图见于 2026 年公开发布自动报告草稿的生产性能数据尚未公开

收益来自公开发布和客户故事页面,而非覆盖所有工作流的标准化第三方基准。

[CE011, CE012, CE016, CE017, CE019, CE020]
FE004: 产品成熟度 / 能力图

能力评分区分了已有强现场部署证据的领域,以及路线图说法超过公开证据深度的领域。

评分是 1–5 有序尺度上的证据支撑分析判断,不是厂商提供的产品评分。

[CE011, CE016, CE017, CE019, CE027, CE028]

5.3 集成、编排与企业级部署

Aidoc 的产品主张一直强调嵌入临床医生已经使用的系统。aiOS、VTE 和主动脉页面都描述了 PACS、EHR 与移动端交付,并在影像分诊之上叠加照护协同。VTE 页面尤其明确:aiOS 摄取多模态数据,把提醒路由到 PACS、EHR 和移动工作流,并在同一运行层中呈现实验室、生命体征、影像复核和 PERT 沟通。主动脉页面对跨科室聊天、移动复核和 EHR 连接的照护触发也提出了类似主张。 FDA 摘要显示了这一设计的监管边界。K252970 是工作流分诊和通知产品,不是诊断软件;用户仍需在 PACS 中阅读完整影像并使用专业判断。这一边界重要,因为它解释了 Aidoc 如何把 AI 深度嵌入运营工作流,同时仍将软件定位为优先级排序和协同层,而不是自主诊断代理。 部署证据显示 Aidoc 可以支持不止一种运行模式。FDA 摘要在 CARE 体部 CT 获批中描述了云环境中的基于 Linux 的服务器。Aidoc 的 AWS 页面营销在 AWS 上符合 HIPAA 的企业级部署;Asklepios 部署则称实施采用安全的云端方法,并配合本地部署 aiOS 平台,以满足 GDPR 和既有放射系统约束。合在一起,公开记录支持的是可配置的云加企业集成模式,而不是单一不可变部署架构。 [CE004, CE006, CE014, CE015, CE018, CE020]

FE003: 关键依赖图

Aidoc 的部署依赖医院影像系统、云 / 安全基础设施、监管纪律和客户治理协同。

具体商业合同条款未公开;该图展示抓取材料中可见的主要依赖类别。

[CE004, CE021, CE023, CE024, CE027, CE035]

5.4 信任、安全、安保与合规

Aidoc 的信任姿态比典型创业 AI 厂商的公开表面成熟得多。质量页面称,Aidoc 运行在通过 MDSAP 和 ISO 13485 认证的 QMS 下,并符合 FDA QSR、EU MDR、ISO 14971 和 IEC 62304。更重要的是,该页面明确重申,公司分诊和通知解决方案不是诊断软件,也不旨在替代临床医生对完整影像的复核。这是对产品边界条件有价值的公开承认。 安全方面,Aidoc 称其对齐 NIST Cybersecurity Framework,并在 AWS 和 Azure 上运行,使用 EDR、加密、SIEM 和 CSPM 等控制。AWS 合作页面还单独把产品营销为符合 HIPAA、面向企业部署。这些说法不能替代客户信任中心审计或经审阅的 SOC 报告,但确实显示 Aidoc 正把自己呈现为医院级、受监管的软件厂商,而不是轻量科室 AI 应用。 BRIDGE 框架在控制和认证之外增加了治理视角。Aidoc 把 BRIDGE 定位为与 NVIDIA、医疗系统和行业伙伴共同制定的路线图,用于有韧性、负责任地采用 AI。它有用,并不是因为证明了技术优势,而是因为它说明 Aidoc 理解:多模型部署成功与治理和运行纪律同样相关,不只是算法准确率。公开缺口在于,BRIDGE 描述的是原则和协作,而不是逐客户接入第三方模型的精确标准。 [CE021, CE022, CE023, CE024, CE027, CE036]

信任 / 质量 / 合规表
控制 / 认证状态范围缺口 / 影响
NIST 网络安全框架已采用公开引用的安全运营框架未按客户环境公开映射每个控制族
AWS 与 Azure 安全栈使用中平台托管和数据保护工具客户级云 / 本地部署拆分未公开
EDR / 加密 / SIEM / CSPM使用中数据保护与网络防御栈公开表述偏描述性,不算可审计证据
MDSAP已认证覆盖医疗器械单一审核计划全球受监管运营的成熟度信号
FDA QSR (21 CFR Part 820)合规受监管产品的器械生命周期控制单凭 FDA QSR 合规不能证明每个工作流的临床有效性
ISO 13485:2016已认证医疗器械质量管理体系有助于赢得医疗系统采购信任
EU MDR 2017/745已认证 / 合规欧洲市场准入框架公开页面没有逐项列出产品的 CE 范围
ISO 14971 和 IEC 62304合规风险管理与医疗软件生命周期纪律有治理信号价值,但不能替代院点级验证

表格汇总 Aidoc 信任页面的公开自述;不能替代客户审阅信任中心文件、BAA、渗透测试摘要或产品级用户指南。

[CE021, CE022, CE023]

5.5 成熟度、生态方向与技术风险

Aidoc 最强的技术信号不只是监管活动,而是在真实医疗系统中的平台成熟度。到 2026 年 4 月下旬,公司和独立媒体都在描述每年分析 60+ million 个患者病例,并部署到近 2,000 家医院。在客户故事层面,Mercy、Hartford 和 Asklepios 提供了直接证据,显示 Aidoc 能从初始协议走到运营部署,而不会被困在单站点试点中。AI 算法、ML 平台、后端、DevOps、云平台和基础设施产品等岗位招聘,也进一步说明公司仍在持续投入企业级产品路线图。 路线图很激进。Aidoc 称 CARE 将扩展到 CT 和 X-ray 工作流,自动生成报告草稿也是近期目标。这些方向契合公司反复使用的“从像素到报告草稿”表述,以及其基础模型能把多年路线图压缩到更短周期的主张。但这些路线图事项仍是前瞻性的,尚未在公开生产部署中得到广泛证明。 因此,主要尽调风险集中在证明深度,而不是表面广度。公开来源没有给出已调和的官方组合数量,CARE 的独立多站点基准仍薄,公司也没有为其描述的 69% 外部模型生态发布详细的第三方模型接入或治理准则。产品看起来已达到企业级且真实存在,但面向市场的叙事走在了公开记录中的参考架构和基准语料之前。 [CE025, CE026, CE027, CE028, CE029, CE030]

路线图 / 发布 / 开发阶段表
日期 / 阶段功能 / 里程碑状态含义来源
2025-0745M+ 患者 / 150+ 家医疗系统 / 开放生态叙事公开披露意味着 Aidoc 从算法供应商转向平台运营方Healthcare IT Today / Fierce
2026-01CARE 全身 CT 综合获批(11 个新增 + 3 个既有适应症)获批验证基础模型能落地受监管分诊工作流Aidoc + FDA / Diagnostic Imaging
2026-0114 项适应症的全身 CT 安全网定位,覆盖 ED 和积压工作流获批 / 已发布把价值从单病种分诊拓到多病种优先级排序Aidoc 发布稿
2026-04Series E 轮 + 新一轮 aiOS 扩张已融资资金将投向更广的 CARE 适应症覆盖和全球 aiOS 部署Aidoc / PRNewswire
2026-04 起从像素到报告草稿的工作流路线图路线图把 Aidoc 从分诊延伸到下游报告辅助Aidoc / PRNewswire
2026-04 起CARE 扩展到 CT 和 X-ray 工作流路线图意味着覆盖面可能超出现有获批集合,走向更广的多模态Aidoc 2026 年 1 月 / 4 月发布稿
2026 年已上线aiOS 托管外部模型;69% 客户运行非 Aidoc 模型按公开表述已上线支撑平台化 / 范围经济逻辑Healthcare IT Today / Fierce

路线图条目都是公开表述,不是经审计的交付承诺。已获批产品、已宣布上线和前瞻性路线图之间,生产成熟度差异很大。

[CE011, CE026, CE027, CE028, CE029, CE030]

5.6 展示

Chapter 06

06客户情况

6.1 规模、地域与客户覆盖

Aidoc 的公开客户叙事有两层:一个宽泛的规模叙事,以及一组更窄的直接具名证明点。在最宽口径上,Healthcare IT Today、Fierce Healthcare、MedCity News 和 HLTH 的 2025 年独立报道重复了公司自报数字:每年在 150+ 个医疗系统中服务超过 45 million 名患者。Aidoc 2026 年 4 月的官方稿和 PRNewswire 稿随后描述了近 2,000 家医院、每年分析超过 60 million 个病例,以及累计分析超过 110 million 个病例。这些数字显然很大,但分母不能互换:医院、医疗系统、年度患者和年度病例衡量的是采用的不同侧面。 具名证明的地域更偏向美国大型一体化交付网络和一家欧洲大型私营运营商。获取的公开来源直接确认了 Hartford HealthCare、Mercy、Sutter Health、WellSpan、Yale New Haven、Renown Health / Carson Tahoe、Temple Health、Advocate Health 和 Asklepios。独立文章还将 Mount Sinai、Northwell 和 University of Miami 列为合作伙伴或用户,但这些更像规模引用,而不是完整记录的部署案例。最稳妥的解读是,Aidoc 拥有广泛平台触达和有意义的参考客户密度,但公开披露客户清单远比 150+ 或 2,000+ 这样的原始规模指标所暗示的要薄。 [CU001, CU002, CU003, CU004, CU018, CU019]

客户分层表
客群买方 / 使用者 / 付款方主要部署模式最强公开证据关键限制
美国大型整合交付网络CIO / 影像 / 服务线负责人;放射科医生和急症护理团队使用;医疗系统付款企业级 aiOS 跨多条服务线部署Hartford、Mercy、Sutter、Advocate各点名系统公开披露的部署深度不均
区域和社区型枢纽—辐射网络卒中或急症护理负责人;ED、放射科和转运团队使用;网络内付款方组合用 AI 协调转诊站点和转运路径Renown Health / Carson Tahoe结果由院点提供,未经独立审计
美国以外的私立医院运营商集团 CMO / IT 负责人;放射科团队使用;运营商付款多院集中式放射科 AI 平台德国 Asklepios公开来源中,Asklepios 之外的国际足迹很薄
专科急症护理项目PERT / 血管 / 神经团队;临床医生使用;医院付款更大企业账户内的科室级工作流增强Yale New Haven 和 Cedars-Sinai VTE 成效服务线证据不能自动证明企业级标准化
高管层数字化转型买方CEO / 首席数字官 / IT 战略委员会;多学科用户使用;医疗系统付款多年期基础设施与治理关系Temple 和 Sutter 高管评论公开采购经济性仍不透明
平台治理 / AI 生态采用者企业 AI 治理团队;临床医生使用托管方案;医疗系统付款aiOS 作为 Aidoc 和外部模型的运行层69% 非 Aidoc 模型主张第三方托管主张来自公司自报

该分层根据公开部署、高管评论和工作流案例研究推断,不来自公司发布的客户分层文件。

[CU003, CU005, CU011, CU014, CU030, CU037]
客户增长 / 采用轨迹表
指标数值日期 / 期间来源置信度含义缺口 / 限制
公开医疗系统覆盖150+ 家医疗系统2025Healthcare IT Today / Fierce / HLTH / MedCity 等来源显示 2026 年口径切换前已有广泛企业触达公司披露,未经审计
公开患者覆盖每年 45M+ 患者2025Healthcare IT Today / Fierce / HLTH 等来源表明规模已越过早期采用阶段患者数不等于签约 logo 或活跃使用者
公开医院覆盖近 2,000 家医院2026Aidoc / PRNewswire确认部署面很广医院数与医疗系统数是不同口径
平台吞吐量每年 60M+ 病例;累计 110M+2026Aidoc / PRNewswire说明有真实运行量分析病例数不等于付费账户数
战略投资方客户4 家医疗系统2025Healthcare IT Today / HLTH / MedCity 等来源支撑买方信心和深层战略绑定各系统投资规模未披露
Hartford 部署深度17 个 FDA 获批算法,覆盖数百万次检查2025 年上线Aidoc + HIT Consultant最强的公开企业级部署证据完整 12 个月扩展结果尚未公开
Mercy 部署深度50 家机构;12+ 个用例截至 2025-02 / 2026 年报道Aidoc 客户故事清晰的多站点标准化证据公开证据来自供应商故事
Asklepios 部署深度28 家医院;每月约 35k 张 CT / X-ray 图像2026Aidoc 公告国际企业级部署证据单一来源客户公告

各行有意混用医院、医疗系统、患者、病例和部署案例,展示 Aidoc 的公开采用叙事会随分母变化。上述指标不应合并成一个“客户数”。

[CU001, CU002, CU003, CU005, CU009, CU011]
FU001: 客户旅程图

Aidoc 公开客户故事呈现一条反复出现的路径:高管背书和参考案例先建立证据,再进入企业 IT 集成、急症工作流上线和多服务线扩张。

旅程阶段来自公开客户故事和高管访谈推断,并非公司发布的销售漏斗或续约流程图。

[CU005, CU009, CU011, CU015, CU030, CU031]
FU002: 采纳 / 部署漏斗

公开证据集从 Aidoc 自报的广泛公司足迹,急剧收窄到少得多的具名企业参考,再收窄到更少的公开量化结果站点。

这是证据漏斗,不是销售漏斗。数值反映抓取来源中的公开证据密度,不是 Aidoc CRM 内部转化率。

[CU002, CU018, CU019, CU024, CU029]

6.2 具名部署与参考客户质量

抓取来源里,最强的具名企业级部署证据来自 Hartford HealthCare。Aidoc 官方上线公告和 HIT Consultant 均称,Hartford 三周内完成初始上线,部署 17 个获 FDA 许可的算法,覆盖数百万次检查,并扩展到放射、心脏、血管、神经和急诊工作流。Mercy 是最清晰的多站点实施参考:Aidoc 公开案例称,aiOS 已在 Mercy 全部 50 家设施运行,同时上线十多个用例,并已分析数百万张影像。 这两家之外,Sutter Health 公开描述了一项多年期战略合作:aiOS 嵌入其照护体系,Sutter 也成为 Aidoc 的西海岸枢纽。Asklepios 给了 Aidoc 重要的欧洲证明点,28 家医院上线,每月分析约 35,000 张 CT 和 X 光影像。Yale New Haven 是量化程度最高的服务线案例之一,因为其肺栓塞响应工作流给出了具体的高级治疗和漏触发激活指标。Renown/Carson Tahoe 与 Temple Health 则说明,Aidoc 的公开证据并不局限于放射科叙事:公司正在向高管买家销售中枢-辐射式卒中编排和企业 AI 操作系统价值。 [CU005, CU006, CU007, CU008, CU009, CU011]

具名客户证据表
客户 / 系统地区 / 客群部署 / 用例生产 vs 试点量化成效限制
Hartford HealthCare康涅狄格 IDN企业级 aiOS 跨放射、心脏、血管、神经和 ED 推出生产上线,附 12 个月扩展计划3 周上线;17 个 FDA 获批算法覆盖数百万次检查供应商和媒体相互印证,但没有公开 ROI / 续约队列
Mercy美国多州医疗系统aiOS 在全部 50 家机构上线,覆盖 12+ 个用例生产分析 2.4M 张图像;标记 249k 张;门诊诊断时间缩短 90%证据来自供应商运营的客户故事
Asklepios德国私立医院运营商急症护理医院集中式放射科 AI 推出生产28 家医院;每月约 35k 张 CT / X-ray 图像单一来源公告;独立印证有限
Sutter Health加州整合交付网络aiOS 部署和联合开发伙伴关系生产 / 扩展阶段该系统服务 3.5M+ 加州居民;平台用于企业级照护体系公开成效指标尚未披露
Yale New Haven Hospital学术 / 转诊医院PE 响应和高级治疗工作流生产案例研究高级治疗使用适当性提升约 40%;发现约 70% 被漏掉的激活服务线案例研究,不是全企业披露
Renown Health / Carson Tahoe枢纽—辐射式卒中网络卒中编排和转运工作流生产站点自报DIDO 减少 32 分钟;LVO 转运快约 30%内部院点数据,非同行评议比较队列
Temple Health学术 / 公立医院高管买方企业级临床 AI 操作系统伙伴关系约 1.5 年后进入生产PACS / EHR 集成和 ROI 纪律有强定性证据公开披露的量化部署深度很少

“生产 vs 试点”根据公开材料措辞推断。各行优先选择有工作流细节的直接具名证据;排除 Mount Sinai、Northwell 或 University of Miami 等较弱的仅媒体点名案例。

[CU005, CU009, CU011, CU012, CU013, CU014]
FU003: 客户证据矩阵

Aidoc 同时拥有具名部署证据和具体工作流指标时,证据质量最强;若公开记录只靠高管背书或媒体点名、缺少部署细节,证据质量就变弱。

评分为 1–3 尺度上的分析判断,3 代表最强公开证据。它衡量证据质量,不衡量实际客户价值。

[CU020, CU021, CU024, CU027, CU029, CU033]

6.3 量化结果与工作流采用证据

Aidoc 最好的公开结果证据来自工作流提速和照护升级指标,而不是合同或收入指标。Yale New Haven 的 PE 响应案例尤其有用,因为它把 Aidoc 关联到临床上重要的结果:合适的高级治疗使用约增加 40%,约 70% 原本会漏掉的潜在激活被识别出来。Cedars-Sinai 也出现在 Aidoc 的 VTE 页面上,是另一个量化站点:PE 照护的治疗启动时间缩短 7 小时(41%),住院时长减少 26%。 Renown Health / Carson Tahoe 补充了卒中网络案例:door-in-door-out 时间缩短 32 分钟,大血管闭塞转运在约六个月后快了大约 30%。Mercy 的公开部署案例给出的不是照护事件百分比,而是运营规模指标,包括已分析 2.4 million 张影像、249,000 项被标记检查,以及门诊诊断时间缩短 90%。WellSpan CEO 也提到一年内分析超过 200,000 个病例。这些都是有意义的工作流证明,但口径并不统一。有的是公司主导案例,有的是站点提供的内部数据,没有一个能替代标准化的跨客户队列。 [CU010, CU013, CU014, CU016, CU017, CU019]

量化成效 / 工作流证据表
站点指标披露值来源类型置信度含义限制
Yale New Haven Hospital高级治疗使用适当性+40%案例研究 / 客户证据说明工作流能提升升级治疗质量,而不只是加快提醒案例研究格式,非经审计研究登记
Yale New Haven Hospital若无 AI 会漏掉的潜在 PERT 激活~70%案例研究 / 客户证据意味着真实工作流中有实质召回收益仅单站点服务线证据
Cedars-SinaiPE 治疗时间快 7 小时(41%)Aidoc VTE 引用若能复现,急症护理运营影响强确认是工作流研究站点,但不能说明企业部署深度
Cedars-Sinai住院时长-26%Aidoc VTE 引用显示财务和吞吐收益所见页面未完全披露研究条件
Renown / Carson Tahoe门进门出时间-32 分钟院点提供的内部数据辐射端到枢纽端转运更快仅内部数据
Renown / Carson TahoeLVO 转运时间快约 30%(133 分钟降至 94 分钟)院点提供的内部数据卒中网络编排更好仅内部数据
Mercy门诊诊断时间-90%Aidoc 客户故事说明多站点规模下,工作流确实加速供应商制作的客户叙事
WellSpan Health一年内分析病例数200,000+引用客户 CEO 的第三方新闻确认有可观的经常性使用未公布基线利用率分母

成效指标来自供应商客户故事、市场案例研究分发和客户高管引述的混合来源。证据点有价值,但不是覆盖已安装客户群的标准化队列报告。

[CU010, CU013, CU014, CU016, CU017, CU019]

6.4 扩张信号、黏性与持久性缺口

公开客户证据显示,Aidoc 往往先以平台层落地,再扩展到更多服务线。Sutter 公告强调多年期基础设施和共同开发,而不是一次性算法。Hartford 公告提到 12 个月内走向更完整的企业级实施。Mercy 自己的案例称,平台标准化让它能从少数孤立 AI 工具扩到更大规模。Healthcare IT Today 和 Fierce 都引用 Aidoc 的说法:69% 客户已经在 aiOS 上运行非 Aidoc 模型,这意味着买方可能越来越把 Aidoc 当作多供应商 AI 资产的治理和编排层。 这种平台黏性叙事方向上利好,但公开持久性证据仍然偏薄。抓取来源没有披露净收入留存、总留存、客户 logo 留存或续约队列数据,也没有公开集中度披露来说明少数企业级系统是否贡献了过大的 ARR 或病例量。因此,虽然公开足迹显示 Aidoc 可以在复杂系统内部扩张,承销客户质量所需的经典 SaaS 持久性指标仍然缺位。 [CU003, CU024, CU025, CU029, CU030, CU031]

留存 / 重复使用 / 满意度表
指标公开数值客群 / 站点置信度含义缺口
净收入留存率未披露公司整体没有公开证据量化扩张耐久性需要队列或账户级 NRR
总留存率 / logo 留存率未披露公司整体公开来源未显示续约稳定性需要 GRR / logo 流失披露
续约率未披露公司整体无法公开判断合同耐久性需要按部署年份划分的续约队列
头部客户集中度未披露公司整体大型企业账户可能具备战略重要性需要前 5 / 前 10 大客户占比
市场口碑背书28 条评论 / 证言;1 个案例研究;1,448 个参考评分FeaturedCustomers显示一定公开客户背书证据平台数据可能经过筛选,不具代表性
采购审查采购软件前需要 ROI 证明Temple Health暗示复购取决于可量化的运营价值Temple 未披露基于成果的续约条款
小型诊所采用阻力前期成本高、集成负担重被点出ITQlick 评论指向大型企业买家之外的适配度较弱属于声誉较弱的目录网站,不是一手客户证据

类似「Undisclosed」的空值条目是有意保留,反映真实的公开信息缺口。不应解读为零值。

[CU020, CU021, CU022, CU023, CU024, CU025]

6.5 客户采用注意事项与反向信号

主要提醒在于,规模说法一部分扎实,一部分依赖叙事。“150+ 家医疗系统”“45M 名患者”“近 2,000 家医院”和“每年 60M 个病例”听起来都像客户数量口径,但衡量的不是同一件事。公开来源证明客户关系的直接程度也不同。Hartford、Mercy、Asklepios、Sutter 和 Yale 是强抓取参考;Cedars-Sinai 是工作流研究站点;Temple 是高管买方证言;媒体列出的 Mount Sinai、Northwell 或 University of Miami,证明力弱于详细案例。 抓取集中最清晰的反向公开来源是 ITQlick,它指出企业级实施昂贵、定价透明度有限、对小型诊所集成负担重。该来源声誉较低,且本身使用 AI 辅助,因此应被视为弱提醒信号,而不是决定性负面。比目录评价更重要的是公开来源没有显示的内容:没有续约队列,没有集中度表,也没有抓取来源直接确认用户要求的一些名称,例如 NYU Langone、Mayo Clinic 或 University of Rochester Medical Center。这些是尽调缺口,不是否证。 [CU020, CU021, CU022, CU023, CU024, CU025]

扩张与集中度风险表
驱动因素 / 风险当前证据影响置信度尽调路径
战略投资者客户Hartford、Mercy、Sutter 和 WellSpan 参与 2025 年融资强烈指向客户背书质量和战略一致性确认投资金额,以及是否存在商业排他或折扣条款
Hartford 企业扩张初次上线后,官方披露 12 个月扩张路径暗示一个大客户内可先落地再扩张索取当前模块数量和上线后使用率曲线
Sutter 多年枢纽角色公开定位为西海岸枢纽和联合开发伙伴比单点解决方案销售更能支撑深度粘性索取已签约服务线范围和续约日期
Mercy 平台标准化50 家院区 / 12+ 个用例被用作反碎片化证据显示 aiOS 能跨院区和工作流扩张索取按院区和模块划分的当前活跃使用分布
开放生态托管据称 69% 客户在 aiOS 上运行非 Aidoc 模型可能提高迁移成本和平台依赖索取第三方模型清单,以及仅 aiOS 账户与全套件账户的客户留存对比
具名客户保密相比宣称的整体覆盖面,公开可引用客户基数偏小尽调难以广泛验证部署质量拿到 NDA 客户名单,并按客户批次安排背调电话
收入集中度未知未披露头部客户占比对大型企业的依赖可能实质影响收入韧性索取客户集中度表和 ARR 瀑布图
指定客户名单未确认抓取来源未直接确认 NYU Langone、Mayo Clinic 和 URMC防止夸大标杆客户覆盖面要求 Aidoc 确认每个指定客户当前能否对外背书

该表把正向扩张驱动因素,与投资者或战略买家在承销客户质量前仍需看到的韧性和集中度证据缺口分开。

[CU003, CU024, CU025, CU029, CU030, CU031]

6.6 证据展项

Chapter 07

07风险

7.1 监管、法律与临床安全风险

Aidoc 最严肃的风险仍是受监管设备负担与临床责任暴露叠加。公司确实有监管资产:其质量页面称已对齐 MDSAP、ISO 13485、EU MDR 和 FDA 质量体系,Aidoc 又在 2026 年 1 月宣布,基于 CARE 基础模型打造的综合分诊产品获得新的 FDA 许可。这些都是有意义的进入壁垒,但也把 Aidoc 推进一个制度环境:模型更新、新适应证和文档标准每年都更关键。FDA 2025 年 1 月生命周期指南和 2025 年 8 月 PCCP 指南都指向更严格的 AI 赋能设备软件证据预期,尤其是模型上线后继续演进时。Aidoc 自己的 BRIDGE 框架承认,信任、合规和工作流纪律是采用的前提;这在方向上让人安心,但不能替代具体产品的变更控制证据。 临床责任是伴随风险。Aidoc 的价值主张是在急诊和急性工作流中更快升级关键发现,因此一次漏检、延迟或带偏差的优先级排序失败,都可能从普通软件缺陷变成患者伤害事件。同行评议文献仍将偏见和泛化性列为医学影像 AI 的未解问题,尤其是产品跨站点、跨人群迁移时。因此,正确的承销姿态不是“Aidoc 不受监管”,而是“Aidoc 已受监管、正在扩张,也因此越来越暴露在把监管和上市后安全做好所需的成本之下”。[CR001, CR002, CR003, CR004, CR005, CR006]

监管 / 法律风险登记表
风险证据可能性严重性当前缓释剩余敞口尽调路径
FDA 模型治理负担2025 年生命周期 + PCCP 指引提高了 AI 功能更新的证据负担与 MDSAP/ISO/MDR 对齐的 QMS 和已获准产品组合新适应症和模型变更敞口高审阅 PCCP、变更日志和近期 FDA 往来函件
新基础模型产品的欧盟合规状态不透明未找到新综合分诊版本的 2026 年 CE-MDR 公开更新公司层面声称已有 MDR 合规产品状态未单独成文前,敞口为中索取公告机构范围和上市发布文件
漏诊或优先级误排发现带来的临床责任分诊会改变护理团队启动流程,并可能影响时间敏感结果急诊工作流验证、客户背书、医生监督高;伤害事件频率低但后果严重获取索赔文件、赔偿责任上限和上市后安全审查流程
网络事件后的 HIPAA / OCR 执法OCR 在 2026 年继续达成勒索软件和解,且和解协议要求监控安全团队、NIST CSF、DPF 认证中高;公开保障细节不完整索取审计证明、BAA 和安全事件指标
算法偏差 / 泛化失败同行评议文献仍在指出影像 AI 的偏差和欠规范风险广泛多站点落地和产品验证计划中高;跨人群外部验证仍有限索取亚组表现,以及按站点 / 模态划分的监控

列举截至 2026-05-20 公开可见的监管和法律问题;未公开的监管反馈、保密质量发现或密封争议可能未被纳入。

[CR001, CR002, CR003, CR004, CR006, CR007]
FR001: 风险热力图
[CR003, CR007, CR015, CR023, CR030, CR035]
FR002: 风险传导图
[CR003, CR007, CR015, CR023, CR030, CR037]

7.2 安全、隐私、报销与采购风险

Aidoc 公开材料指向一个严肃但只部分披露的控制环境。安全页面提到 NIST CSF 和 AWS 托管,数据传输通知称 Aidoc 已获得 EU-U.S.、UK 和 Swiss DPF 框架认证。这是有用的基线证据,但不同于看到当前 SOC 2、HITRUST、渗透测试、事件响应或 BAA 细节。在医疗采购中,保障细节缺失与已知弱点几乎同样重要,因为它会拉长安全审查,也让数据泄露下行更难定价。HHS OCR 2026 年勒索软件和解案说明,HIPAA 执法是持续且运营层面的,不是理论风险。 报销风险又放大采购风险。独立放射科来源和 ACR 都指出,AI 报销仍落后于获 FDA 许可工具的数量,因为大多数影像 AI 产品仍绑定 Category III 跟踪代码,而不是可持续、与支付挂钩的 Category I 经济性。即使没有直接报销,Aidoc 仍可像 Advocate Health 和 Novant 所显示的那样靠 ROI 销售;但这让每笔企业销售更依赖本地预算归属、工作流重设计,以及可衡量吞吐量或结果改善的证明。这条路能走通,但与有清晰报销支撑的护理标准相比,采用更慢、黏性更强,也更受 CFO 审视影响。[CR013, CR014, CR015, CR016, CR017, CR026]

运营 / 质量 / 安全风险登记表
失效模式证据可能性严重性缓释成熟度剩余敞口未解决缺口
跨站点性能漂移学术文献和全球部署都指向领域迁移风险重大需要亚组和站点级监控输出
安全保障深度未披露公开材料提到 NIST/DPF,但未披露 SOC 2 或 HITRUST 细节重大需要审计函和渗透测试摘要
快速铺开带来的支持负担Asklepios、Advocate、Sol 和 Isala 部署同步扩张重大需要实施人员配置和 SLA 数据
病例量扩张快于监控能力已分析 100M+ 患者病例,抬高上市后监测负担重大需要事件率、人工改判指标和质量审查节奏

剩余严重性反映现有缓释措施失效或规模化不均时的运营后果。

[CR010, CR013, CR015, CR028, CR032]

7.3 合作伙伴、平台与竞争风险

Aidoc 的运营模式与外部平台深度交织。这部分是优势:公司称 aiOS 不绑定供应商,已集成 PACS/VNA/EHR 工作流,并可在 Epic App Orchard 中使用,这正是医院想要的工作流可信度。Aidoc 还与 AWS 合作,有多年期投资支持 CARE 基础模型开发。这些关系降低商业化摩擦,让 Aidoc 比狭窄算法供应商更像企业级公司。 同样的关系也制造集中度风险。如果 Epic 调整姿态,OEM 或转售渠道表现不佳,或 AWS 经济性变化,Aidoc 的部署动作会很快削弱。既有技术栈内部的竞争强度也在上升:Microsoft 面向 Epic 环境销售 Dragon 和 Azure AI,Oracle 推出面向文档和工作流的临床 AI 智能体,Epic 自身也把 AI 定位为嵌入其软件全线。若医院足够看重急性影像分诊和跨团队编排,愿意购买专门层,Aidoc 仍能赢;但这是定价权问题,不是保证成立的护城河。实际结论是,Aidoc 的伙伴姿态是重要缓释项,但应先被视为暴露地图,其次才是护城河。[CR018, CR019, CR020, CR021, CR022, CR023]

合作伙伴 / 依赖风险登记表
依赖交易对手角色集中度信号失效场景严重性缓释剩余敞口
云 + 战略算力伙伴AWS托管平台,并出资支持 CARE 开发多年战略合作和 AWS 托管平台价格、政策或技术扰动拖慢模型路线图战略关系和企业收入增长
工作流入口Epic / App OrchardEHR 工作流分发和互操作Aidoc 强调独特 App Orchard 身份Epic 收紧访问,或打包足够多原生 AI,削弱 Aidoc 价值厂商中立集成和客户证据
竞争性工作流 AIMicrosoft / Dragon / AzureEpic 场景内的环境式和生成式 AIMicrosoft 推广针对 Epic 优化的 AI 工作流临床医生在单独影像 AI 预算获批前,先采用既有 AI 技术栈Aidoc 聚焦急性影像分诊和编排中高
EHR 原生自动化Oracle Health临床文档和工作流自动化Oracle 广泛营销临床 AI 智能体医院把 EHR 厂商 AI 作为标准配置,而不是采用叠加式单点方案Aidoc 在影像工作流里扎得更深
OEM / 转售渠道PACS 和 OEM 伙伴分发和集成提速Aidoc 提及转售伙伴,但未给出完整名单或集中度渠道冲突或转售放缓会推高 CAC 和实施时间直销加客户证据

按交易对手对产品交付、工作流访问或利润率的直接影响程度排序。

[CR018, CR019, CR020, CR021, CR022, CR023]
FR003: 依赖关系图
[CR018, CR019, CR020, CR021, CR022, CR023]

7.4 执行、领导力与投资否决条件

Aidoc 的执行风险随规模上升。到 2026 年 1 月,公司称 aiOS 已超过 100 million 个已分析患者病例;到 2 月和 4 月,公司公开宣传在德国、荷兰、南加州和 Advocate Health 的多站点部署。这是产品市场匹配证据,但也意味着实施质量、上市后监测和客户成功纪律比早期算法阶段更重要。领导力方面,Aidoc 在 2026 年显著补强班底,任命前 AMA 主席 Jesse Ehrenfeld 为首席医疗官;领导层页面也显示,其高管团队比纯粹以创始人为中心的初创公司更成熟。即便如此,Elad Walach 仍是公司公共叙事和融资姿态的中心,因此关键人风险降低了,但没有消失。 最有用的投资者姿态,是提前定义否决条件。若 Aidoc 遭遇具名患者伤害或执法事件、失去特权互操作访问、因报销和预算归属持续模糊而无法转化试点,或无法让模型变更治理与 FDA 预期保持一致,投资逻辑就应重新评估。公司 2026 年大额 Series E 融资缓释了现金跑道风险,但也意味着未来融资和退出预期会被放到更高证明门槛下评判。[CR028, CR032, CR033, CR034, CR035, CR036]

人员 / 执行风险登记表
角色 / 职能依赖或缺口可能性严重性现有缓释剩余敞口尽调路径
首席医疗负责人此前 CMO 可见度有缺口,Jesse Ehrenfeld 现已补位中低高知名度临床领袖于 2026 年加入审阅决策权和安全治理章程
创始人 / CEO 中心化Elad Walach 仍是产品和融资叙事的核心公开面孔更广管理梯队现已公开列出索取继任和授权运营负责人计划
实施与客户成功2025-2026 年大规模铺开增加服务负担企业平台和重复客户斩获中高获取部署人员比例和 SLA 指标
上市后质量运营100M+ 已分析病例抬高监测工作量QMS 和全球落地纪律中高索取安全审查委员会输出和升级处理指标

2026 年领导层广度改善,但执行复杂度也同步上升。

[CR028, CR032, CR033, CR034]
缓释与否决标准表
风险可监控触发器阈值 / 事件行动含义
监管负担FDA 沟通或申报节奏核心模型更新出现意外新申报,或重大 PCCP 缺陷暂停承销,直到治理材料审阅完毕
临床责任具名医院发生与漏诊或优先级误排发现挂钩的伤害事件任何引发监管或法院审查的公开患者伤害事件升级为投资论点破裂审查
安全 / 隐私重大事件或外部审计缺口确认的数据泄露、OCR 调查,或影响客户的安全审计失败冻结新增资本,等待整改证据
支付 / 采购试点转生产转化率预算或支付路径不清,导致大客户试点后未转生产下调增长假设并重估估值
伙伴依赖Epic/App Orchard 或 OEM 访问变化失去特权互操作、出现实质 API 限制,或转售受扰视为护城河受损
执行多个企业部署未达实施 SLA铺开中延误升级或客户明显反弹上调服务成本和流失假设

阈值是实用的投资监控触发器,而非绝对经营预测。

[CR030, CR031, CR040, CR041, CR042, CR043]
Chapter 08

08估值

8.1 建议与估值立场

Aidoc 看起来是有真实牵引力的真实品类参与者,但公开证据仍不足以支持可投资的价格判断。公司可以拿出新的 FDA 许可、具名企业部署、超过 100 million 个已分析患者病例,以及 AWS、Goldman Sachs Alternatives、General Catalyst、SoftBank Vision Fund 2 和 NVentures 的战略支持。这足以排除 Aidoc 只是概念股故事的轻率看法;但还不足以让投资者盲目支付溢价私募估值。 核心问题是估值不透明。2024 年后公开融资报道确认了更多资本和更强投资人支持,但没有给出清晰的当前投后估值。这意味着公众无法把今天隐含的私募价格换算成收入倍数、经利润率调整的倍数或经优先权调整的回报。我的立场因此是继续研究 / 观察。若内部指标强,Aidoc 可能值得溢价;但今天的公开证据支持好奇,不支持确信。由于 2024-05-20 之后的当前估值未披露,Aidoc 的独角兽身份应视为未确认标签,而非已验证事实。[CV003, CV011, CV012, CV038, CV039, CV043]

建议摘要表
维度评估置信度估值立场决策含义
整体建议继续研究 / 跟踪未知到偏高当前估值和收入质量未披露前,不要按溢价入场承销
当前公开价格可见度2024-05-20 后未披露Unknown把当前私募市场价格视为证据缺口,而非事实
公开牵引质量有意义但不完整支撑企业落地和 FDA 进展足以支撑尽调,但不能支撑盲目接受价格
可比公司支撑混合受公开可比公司约束公开医疗 IT 可比公司估值远低于稀缺性溢价的私募叙事
下行保护公开证据显示较弱风险高优先权结构和稀释仍需直接尽调

公开证据支持方向性建议,但当前轮条款未披露,无法给出精确入场倍数。

[CV003, CV011, CV012, CV039, CV043, CV044]
论点 / 反论点表
维度论点理由反向逻辑什么会改变判断
商业牵引力具名医疗系统上线和 100M 分析病例显示采用动能真实上线数量不等于已披露付费收入或持久扩张按队列拿出已签 ARR、付费使用率和续约数据
战略背书Goldman、AWS、General Catalyst、SoftBank 和 NVentures 为这个品类背书战略资本可能放大价格不透明,而不是消除它披露当前投后估值和董事会估值备忘录
监管证明新 FDA 许可降低「科研项目」风险监管进展不能证明单位经济或达到退出质量的收入按获批产品展示收入贡献和利润率画像
私有可比公司Abridge 和 Viz.ai 说明头部医疗 AI 厂商可能拿到溢价估值PathAI 和 Paige 说明并非每家医疗 AI 公司都能维持高溢价的独立结局说明 Aidoc 为什么能越过溢价门槛,而不是走向出售或估值压缩
市场环境后期市场挑剔但仍在运转,赢家仍能拿到奖励Rock Health 和 Cooley 都提到资本集中、条款继续挑剔拿出明显优于同行的增长和效率画像
独角兽标签如果近期未披露轮次提价,Aidoc 可能已高于这一象征性门槛2024-05-20 之后没有公开来源证明当前价格达到或超过 $1B拿出当前新股轮估值披露或经审计的老股交易参考

核心分歧不在于 Aidoc 是否真实,而在于其未披露的当前估值能否被公开证据支撑。

[CV009, CV011, CV012, CV034, CV035, CV036]
FV001: 投资建议逻辑与尽调关口
[CV039, CV043, CV045, CV046]

8.2 融资历史与公开来源真正确认的内容

Aidoc 最后一个完全公开、披露金额的轮次,是 2022 年 6 月宣布的 $110 million Series D。该公告清楚披露了轮次规模和累计融资额,但没有披露投后估值。此后的公开证据不是更透明,而是更不透明。Aidoc 2026 年 4 月 Series E 新闻稿披露新增 $150 million 融资,累计融资超过 $500 million。2025-2026 年独立媒体报道也描述了追加或更高融资,但仍没有锁定当前投后价值。 这种不对称很重要。投资者常假设后续轮次会自动澄清价格发现;Aidoc 的公开记录恰好相反。它证明资本继续进入、投资人阵容改善、公司仍在推进;但没有证明新投资人实际支付的价格、股权结构如何变化,或标题估值能否经受公开市场或战略收购方的现实检验。换句话说,Aidoc 的融资历史现在是强赞助证据,却是弱价格证据。[CV001, CV002, CV003, CV004, CV009, CV010]

8.3 可比估值锚点

Aidoc 最干净的公开锚点,不是另一个不透明私募轮,而是高溢价公开 AI 倍数与主流医疗 IT 倍数之间的差距。按引用的 2026 年锚点,Tempus 是溢价异常值,市值 / 收入约 9.0x。RadNet 和 GE HealthCare 更接近 2x-3x,Phreesia 和 Health Catalyst 则低得多。信息很直接:公开市场愿意为差异化医疗 AI 和数据规模付溢价,但只有当收入规模和质量可见时,溢价才会变得持久。 私募可比公司给出类似但更细的故事。Viz.ai 和 Abridge 显示,品类领先的医疗 AI 公司可以获得丰厚融资支持。PathAI 2026 年 Roche 交易和 Paige 2025 年出售给 Tempus 说明,并非每个临床可信的 AI 资产都会像独立平台那样定价。对 Aidoc 来说,这意味着上行情景真实存在,但证明负担也真实存在。没有披露经济指标时,理性做法是把可能价值放在这些可比区间内框定,而不是把私募稀缺性当成自证。[CV013, CV014, CV015, CV017, CV020, CV023]

可比估值表
可比样本收入 / 规模锚点估值锚点隐含倍数或价格为什么可比局限
Tempus AI2025-09-30 收入约 $904.6M市值约 $8.17B~9.0x最接近的上市高 AI 含量医疗数据平台溢价锚点基因组学和精准医疗范围比 Aidoc 更宽
RadNet2025-09-30 收入约 $1.492B市值约 $4.19B~2.8x具备运营规模的上市影像服务锚点服务占比高,不是软件式毛利率画像
GE HealthCare2025-09-30 收入约 $14.927B市值约 $28.01B~1.9x大型影像设备和工作流锚点业务太多元、硬件太重,难以直接按软件可比
Phreesia2025-10-31 收入约 $353.5M市值约 $0.56B~1.6x医疗工作流软件,提供公开市场校准不是影像或急性护理 AI 公司
Health Catalyst2025-09-30 收入约 $236.5M市值约 $93.84M~0.4x上市医疗 IT 极低端倍数锚点重整案例,不是健康的溢价同行
Viz.ai1,000+ 家医院使用平台2022 年私募估值 $1.2B$1.2B 投后最接近的私有急性护理影像 AI 先例估值较旧,公开收入透明度有限
Abridge150+ 家企业级医疗系统2025 年 Series E 轮融资 $300M溢价私募融资轮说明后期医疗 AI 仍能拿到高价文档 AI 与影像分诊的经济性不同
PathAI2026 年战略出售给 RocheUSD 750M 首付款 + 最高 USD 300M 里程碑款战略并购锚点病理 AI 退出先例,可供参考病理工作流不同于影像分诊
Paige / Tempus2025 年出售给 Tempus收购价 USD 81.25M小型战略补强收购提醒:并非每项医疗 AI 资产都能拿到独角兽结局单一病理资产业务,不是广泛医院平台

枚举样本经过筛选,但有意混合了与 Aidoc 品类相关的上市和私有医疗 AI 或影像锚点。

[CV013, CV014, CV015, CV017, CV020, CV023]
FV002: 估值敏感性
[CV033, CV040, CV041, CV042, CV047, CV048]

8.4 乐观、基准与悲观估值区间

由于当前私募价格未披露,正确的情景框架应看退出价值,而不是当前内在价值。乐观情景下,Aidoc 把现有企业牵引力和基础模型叙事转化为数亿美元经常性收入,同时保住医疗 AI 溢价倍数。这可以支撑中等个位数十亿美元退出区间。基准情景下,Aidoc 证明自己是有价值但并不例外的医疗工作流平台,指向较低个位数十亿美元结果。悲观情景下,报销摩擦、EHR 原生竞争或当前部署的弱商业化,把 Aidoc 推向 1x-4x 的公开医疗 IT 走廊。 情景逻辑并不是 Aidoc 必须失败才配得上更低估值;而是即便是好公司,如果资本市场正常化快于运营证据到来,也可能跑输溢价私募定价。Rock Health 和 Cooley 都把 2025-2026 年数字健康市场描述为集中且选择性强,这一点尤其成立。正确问题不是“Aidoc 能值很多钱吗?”而是“在溢价标记站得住之前,Aidoc 必须跨过怎样的收入和质量门槛?”[CV033, CV034, CV035, CV036, CV040, CV041]

乐观 / 基准 / 悲观情景表
情景经营假设隐含退出估值估值逻辑关键风险概率信号
乐观收入大约达到 $450M-$700M,扩张强劲,AI 定位享受溢价USD 4.5B-7.0B基于医疗 AI 溢价收入基数,按 8x-10x规模化执行、监管、报销需要 Tempus / Abridge 式收入可信度,并守住持续溢价叙事
基准收入大约达到 $300M-$450M,扩张扎实但并非顶尖USD 1.8B-3.4B基于上市医疗 IT 溢价区间,按 5x-7x竞争与采购拖累对应一家好公司,但定价更接近上市可比公司
悲观收入停在约 $150M-$250M,市场对不透明失去耐心USD 0.6B-1.4B压缩后的上市医疗 IT 区间,按 2x-4x预算摩擦、报销、EHR 原生竞争对应持平 / 下调轮融资路径和估值重置

情景是退出价值案例,不是当前价值断言,因为 Aidoc 当前私募价格并未公开披露。

[CV040, CV041, CV042, CV047, CV048, CV049]
FV003: 估值 / 回报区间
[CV047, CV048, CV049]

8.5 尽调问题与投资逻辑失效触发器

最重要的尽调项最简单:当前价格是多少?没有这个数字,即便最好的可比分析也只是合理性测试。第二个阻碍是收入质量——ARR、利润率、留存、付费使用转化和优先股堆叠。如果这些指标强,Aidoc 可能合理享有溢价;如果表现平庸,公开可比区间会迅速压缩估值论证。 这直接引出投资逻辑失效触发器。在 2025-2026 年不透明背景下,平轮或下轮会是红旗,说明早期私募标记跑在基本面前面。无法把标志性部署转化为付费扩张,会以更慢速度讲同一个故事。如果报销或既有工作流供应商继续阻碍医院为 Aidoc 分配持久预算,公司看起来就更像有用功能层,而非高溢价独立平台。在这些问题回答前,正确收尾姿态是有纪律的好奇,而不是急切。[CV037, CV039, CV043, CV045, CV046]

投资逻辑破裂与否决触发表
触发点阈值对投资逻辑的传导行动含义
不利融资重置相对未披露的 2025-2026 定价背景,任何持平或下调轮融资表明当前私募标记跑在基本经济性前面从上市可比区间重新承销,不再沿用此前私募标记
收入质量偏弱已签 ARR、扩张或毛利率数据撑不起 AI 溢价叙事稀缺性溢价的支付理由被打穿在溢价定价下,立场从继续研究转向回避
上线但未变现企业部署没有转化为付费经常性扩张牵引力变成虚荣指标,不再创造价值下调收入假设和退出区间
报销瓶颈由于支付路径不清,医院无法把 Aidoc 资助到试点之外限制采用速度和定价权压缩基准和乐观估值情景
EHR 原生玩家侵蚀既有工作流厂商吸收这一预算品类削弱差异化和独立价值视为护城河受损,并偏向战略出售
激进优先股堆叠最新轮包含投资人保护条款,使新资金经济性处于劣后表面估值不再反映投资人结果质量暂停推进,先要求完成股权结构表尽调

这些是潜在投资人的投资逻辑破裂触发点,不表示触发点已经发生。

[CV037, CV039, CV042, CV045, CV046]
最终尽调要求表
主题缺失证据重要性负责人 / 尽调路径
当前投后估值精确轮次价格和股权结构表标记没有它,就无法守住入场纪律要求提供已签投资条款清单和董事会材料
收入质量ARR、GAAP 收入、毛利率、NRR 和队列扩张用来把价格换算成倍数,并判断持续性要求提供 CFO 认证的经营指标包
优先股堆叠清算优先权、参与权、反稀释和期权池扩张决定表面估值能否转化为投资人结果与律师一起审查融资文件
付费部署转化按主要客户列出合同价值、使用率、续约和扩张区分明星客户 logo 和持久变现要求提供客户队列商业仪表盘
按工作流拆分的单位经济按用例列出实施成本、支持成本、云成本和回本周期用来检验溢价软件倍数是否站得住要求提供产品线贡献利润率模型
监管路线图待提交事项、PCCP 范围和上市后监测节奏监管负担可能拖慢发布并稀释倍数审查 RA/QA 路线图和往来记录

这些要求是从方向性判断走向可投资估值判断所需的最低资料包。

[CV039, CV043, CV045, CV046]
FV004: 投资 KPI
[CV001, CV003, CV011, CV020, CV023, CV026]

免责声明

本报告只是基于公开证据的尽调快照,不构成投资建议。关键财务、法律、技术和合同事实仍未公开;任何投资决策前,都应直接向管理层核验,并查阅一手文件。

证据索引

结论
编号陈述可信度来源
CO001 Aidoc was founded in 2016. SO011, SO015
CO002 Aidoc’s three verified co-founders are Elad Walach, Michael Braginsky, and Guy Reiner. SO002, SO003, SO004, SO011
CO003 Elad Walach is Aidoc’s co-founder and CEO. SO002, SO011
CO004 Michael Braginsky is Aidoc’s co-founder and CTO. SO004, SO011
CO005 Guy Reiner is Aidoc’s co-founder, chief architect, and general manager of the Tel Aviv branch. SO003, SO011
CO006 Aidoc retains a material Israeli operating footprint through its Tel Aviv branch even as recent financing announcements are datelined from New York. SO003, SO013, SO014
CO007 Fetched primary sources support a New York-based U.S. operating presence but do not surface a dedicated corporate headquarters page that cleanly resolves New York versus Tel Aviv as the sole headquarters. SO013, SO014, SO003
CO008 Aidoc describes itself as a clinical AI company focused on helping healthcare teams optimize patient treatment and economic value. SO001
CO009 aiOS is Aidoc’s clinical AI platform for running, orchestrating, and governing clinical AI across a health system. SO005, SO013
CO010 aiOS integrates with PACS, EHR, mobile, and care tools to prioritize urgent and nonacute findings inside existing workflows. SO005, SO008
CO011 CARE is Aidoc’s clinical-grade foundation model trained on real-world multimodal data. SO006, SO010
CO012 Aidoc says CARE already powers FDA-cleared applications across multiple clinical domains. SO006, SO010, SO017
CO013 Aidoc’s current product scope extends beyond radiology triage into care coordination and patient management workflows. SO007, SO019
CO014 Aidoc’s radiology portfolio spans neurovascular, aortic, cardiology, venous thromboembolism, chest, abdomen, and partner-delivered add-ons inside a centralized widget workflow. SO008
CO015 Aidoc’s February 2022 Series D raised $110 million and brought cumulative funding to $250 million at that time. SO009
CO016 The 2022 Series D was co-led by TCV and Alpha Intelligence Capital, with participation from CDIB Capital. SO009
CO017 Aidoc’s July 2025 CARE financing raised $150 million plus a $40 million revolving credit facility and brought cumulative funding to $370 million. SO010, SO011, SO012
CO018 The 2025 financing was led by General Catalyst and Square Peg with NVentures and several major U.S. health systems participating. SO010, SO011, SO012
CO019 Aidoc’s April 2026 Series E raised $150 million led by Growth Equity at Goldman Sachs Alternatives. SO013, SO014, SO015
CO020 General Catalyst, SoftBank Vision Fund 2, and NVentures participated in the 2026 Series E round. SO013, SO014, SO015
CO021 Aidoc’s 2026 financing materials say cumulative funding is now over $500 million. SO013, SO014, SO015
CO022 Aidoc’s 2026 Series E materials say the company analyzes more than 60 million patient cases annually. SO013, SO014, SO015
CO023 Goldman Sachs’ April 2026 investor release says Aidoc’s technology has analyzed more than 110 million patient cases and supports approximately 70 million patients each year. SO014, SO013
CO024 Aidoc’s July 2025 CARE financing release said the company supported more than 45 million patients annually across 150+ health systems and projected 100 million in three years. SO010
CO025 Aidoc’s 2026 Series E materials say the platform is deployed across nearly 2,000 hospitals worldwide. SO013, SO014, SO015
CO026 Globes reported in July 2025 that Aidoc systems were installed in about 2,000 hospitals, most of them paying customers. SO012
CO027 Asklepios completed an Aidoc aiOS rollout across 28 hospitals and about 35,000 CT and X-ray images per month. SO018
CO028 Hartford HealthCare implemented Aidoc’s aiOS with 17 FDA-cleared algorithms across millions of annual patient exams and reached initial go-live in three weeks. SO019, SO005
CO029 Aidoc’s comprehensive body CT triage clearance combines 11 newly cleared indications with three previously cleared indications into one workflow. SO017
CO030 Aidoc’s February 2025 landmark clearance applied foundation-model technology to a rib-fractures triage solution and framed the device as the first QFM SaMD of its type. SO016
CO031 FDA clearance K231631 cleared BriefCase-Quantification for coronary artery calcification quantification in November 2023. SO021, SO020
CO032 Aidoc’s CAC-01 model card says the coronary artery calcification model was trained, tuned, and validated on more than 21,000 scans. SO020, SO021
CO033 FDA clearance K213721 cleared BriefCase for brain aneurysm triage and notification in March 2022. SO022, SO023
CO034 Aidoc’s disclosed quality and security stack includes ISO 13485 MDSAP, EU MDR, 21 CFR Part 820 compliance, ISO 27001/27017/27018/27799, SOC 2 Type 2, Cyber Essentials, and C5. SO020
CO035 Aidoc says its AI Monitoring team tracks performance 24/7 to mitigate model drift. SO020
CO036 Aidoc’s model card acknowledges that race, ethnicity, language, sexual orientation, gender identity, and social determinants are not accessible in DICOM images, so bias mitigation relies on proxies and post-deployment validation. SO020
CO037 Aidoc’s clinical compendium claims more than 100 peer-reviewed publications or abstract and conference presentations. SO024, SO020
CO038 Aidoc’s 2025 financing release publicly named Hartford HealthCare, Mercy, Sutter Health, WellSpan Health, Mount Sinai Health System, Yale New Haven Health System, Northwell Health, University of Miami Health System, and Temple Health as partner health systems. SO010
CO039 Official Aidoc pages reviewed during this run did not substantiate a “170+ AI models” count; they support broad algorithm coverage and multiple FDA-cleared solutions instead. SO005, SO006, SO008, SO017
CO040 Post-2024-05-20 sources fetched for this chapter do not disclose a valuation or confirm private unicorn status, so Aidoc’s current unicorn status remains unverified despite the company’s large capital base. SO012, SO013, SO014
CO041 The 2025 CARE financing release says Aidoc planned to invest over $150 million in the coming years through strategic initiatives with NVIDIA and AWS. SO010
CO042 Aidoc’s 2025 CARE financing release says 69% of its customers were already running non-Aidoc models on aiOS. SO010
CO043 The names Cedars-Sinai, Mayo Clinic, and NYU Langone were not independently confirmed in the fetched primary sources for this run even though third-party aggregators sometimes associate them with Aidoc. SO010, SO018, SO019
CO044 A May 2026 real-world study summary said Aidoc’s pulmonary embolism algorithm matched radiologist interpretations in 97.8% of 32,501 scans over 18 months. SO026
CO045 The same study summary said Aidoc missed 15% of confirmed pulmonary embolism cases and radiologists were correct in 89.8% of cases where the algorithm missed a PE they identified. SO026
CO046 The reimbursement literature Aidoc operates against says generalist radiology AI does not fit existing reimbursement frameworks cleanly, which makes enterprise ROI more dependent on operational savings than direct payment codes. SO025, SO012
CO047 Aidoc did not disclose revenue figures in the fetched 2025 Globes interview or the 2026 Series E materials. SO012, SO013, SO014
CO048 A precise public headcount was not confirmed in the fetched sources for this run. SO001, SO013, SO014
CO049 Aidoc positions aiOS as an open enterprise platform rather than a single-vendor point solution, with one integration intended to support every workflow. SO005, SO010
CM001 Aidoc’s core market is enterprise radiology AI and related care-coordination workflows rather than a general-purpose EHR or hospital-operations platform. SM001, SM002
CM002 Aidoc’s radiology positioning emphasizes triage, prioritization, follow-up activation, and deep integration with PACS, EHR, scheduling, and reporting systems. SM001
CM003 Aidoc’s care-coordination positioning emphasizes multidisciplinary activation and real-time routing of patients beyond the radiology reading room. SM002, SM004
CM004 Asklepios’ 28-hospital rollout shows Aidoc can operate as a group-wide radiology platform rather than a single-site pilot. SM003
CM005 Hartford HealthCare’s three-week initial go-live shows that Aidoc’s enterprise deployment story is built around operational speed as well as clinical breadth. SM004
CM006 The American Hospital Association says there are roughly 6,100 hospitals and 907,216 staffed beds in the United States. SM012
CM007 CMS says national health expenditures reached $5.3 trillion in 2024, including $1.6347 trillion of hospital spending and $1.1097 trillion of physician and clinical services spending. SM011
CM008 NCI estimates 2,041,910 new U.S. cancer cases in 2025 and records U.S. cancer-care expenditures of $208.9 billion in 2020. SM014
CM009 WHO says cancer caused nearly 10 million deaths in 2022 and lung cancer alone represented 2.5 million new cases. SM013
CM010 Aidoc’s relevant market excludes hardware imaging systems, billing systems, and generic EHR platforms even though those systems remain integration dependencies. SM001, SM002, SM026
CM011 Radiology remains the densest category on the FDA AI-enabled medical-device roster, which helps explain why Aidoc competes in a crowded but validated regulatory field. SM005, SM026
CM012 FDA’s AI/ML SaMD Action Plan treats radiological imaging workflow automation as a core AI device domain, reinforcing Aidoc’s category legitimacy. SM006, SM005
CM013 FDA’s PCCP draft guidance shows that adaptive medical AI still requires formal change-control planning, which raises lifecycle cost and slows model iteration. SM007, SM006
CM014 The 510(k) framework requires substantial equivalence to a predicate device, which favors incumbents once an indication has an accepted predicate. SM008, SM009
CM015 Viz.ai’s original LVO workflow entered through a De Novo review rather than a 510(k), illustrating how first movers can create the predicate base that later vendors use. SM009, SM008
CM016 Aidoc’s current go-to-market narrative has shifted from single-condition tools toward broader enterprise clinical AI deployed across entire health systems. SM023, SM024
CM017 Aidoc’s July 2025 financing release said the company supported more than 45 million patients annually across 150+ health systems. SM025, SM023
CM018 Aidoc’s April 2026 financing materials said the platform was deployed across nearly 2,000 hospitals and handled more than 60 million patient cases annually. SM024, SM023
CM019 A company claiming nearly 2,000 hospitals and enterprise rollouts at Hartford and Asklepios should be treated as being beyond pilot-stage commercial proof even if pricing is undisclosed. SM003, SM004, SM024
CM020 Neiman HPI projects that the present radiologist shortage will persist through 2055 unless workforce supply rises faster or imaging utilization per person falls. SM017, SM018
CM021 Neiman HPI estimates imaging utilization will rise 16.9% to 26.9% by 2055 even without assuming worsening per-person use. SM017
CM022 Neiman HPI says radiologist attrition has run 50% higher since 2020 than before COVID, materially tightening workforce supply. SM017
CM023 AAMC projects a total physician shortage of 13,500 to 86,000 by 2036, reinforcing the broader labor scarcity that makes clinical workflow AI more attractive. SM016, SM015
CM024 ACR’s 2026 workforce update says economic and regulatory pressures are making radiology practice harder, supporting the case for workflow automation and triage tools. SM018, SM017
CM025 A real-world Northwell Health study summary said Aidoc’s pulmonary embolism algorithm matched radiologist interpretations in 97.8% of 32,501 CTPAs over 18 months. SM021
CM026 The same study summary said the algorithm missed 15% of confirmed pulmonary embolism cases and that radiologists were favored in most discordant adjudications. SM021
CM027 Peer-reviewed reimbursement literature says current reimbursement frameworks fit narrow radiology AI poorly and fit generalist radiology AI even less cleanly. SM019, SM020
CM028 Radiology Business reported that only two Category I CPT codes existed for newer imaging AI heading into 2026 despite hundreds of FDA-cleared algorithms. SM022, SM005
CM029 Most fracture, incidental-finding, and lung-nodule AI tools are unlikely to get their own paid CPT codes because those findings are already part of the reimbursed imaging read. SM022, SM019
CM030 For most imaging AI tools, ROI therefore comes from efficiency, fewer repeats, or better downstream patient management rather than direct reimbursement. SM022, SM019
CM031 FDA clearance and clinical efficacy do not automatically create payment coverage. SM019, SM022
CM032 Aidoc’s core buyers are health systems and radiology-led enterprise service lines rather than direct-to-consumer patients or standalone payers. SM001, SM002, SM004
CM033 Budget ownership for Aidoc-like deployments likely sits with radiology leadership, CIO/CMIO, innovation, or enterprise operations depending on whether the purchase is a departmental or system-wide rollout. SM003, SM004, SM001
CM034 Asklepios explicitly links Aidoc adoption to radiology shortages, smaller-location coverage, and night-and-weekend support. SM003
CM035 Hartford frames Aidoc as a cross-department AI governance and care-delay reduction tool spanning radiology, cardiology, vascular, neurology, and emergency departments. SM004
CM036 Aidoc’s own differentiation argument is deep workflow integration plus multi-algorithm orchestration rather than single-algorithm accuracy alone. SM001, SM023
CM037 MarketsandMarkets sizes the global radiology AI market at $0.76 billion in 2025 and $2.27 billion in 2030, a 24.5% CAGR. SM026
CM038 Emergen Research says hospitals represented about 55% of the global radiology AI market in 2024. SM027
CM039 MarketsandMarkets says hospitals are the largest end-user segment in radiology AI and CT is the leading modality segment. SM026
CM040 Aidoc’s care-coordination layer broadens the company’s economic relevance from radiologist triage into downstream specialty activation and follow-up management. SM002, SM004
CM041 Lifecycle governance requirements for adaptive medical AI create barriers to entry that favor vendors with mature monitoring, quality, and integration infrastructure. SM006, SM007, SM023
CM042 Aidoc’s raw model-card disclosure says the company monitors model performance continuously and maintains drift controls, supporting its enterprise-governance pitch. SM023
CM043 Aidoc’s model-card disclosure also shows that some demographic fairness variables are absent from the underlying DICOM stream, preserving generalization and bias limits even with mature governance. SM023
CM044 Aidoc’s platform narrative does not remove the need for indication-specific regulatory evidence; products such as K231631 still have to clear distinct regulatory endpoints. SM010, SM008
CM045 Aidoc’s own 2026 Series E copy says the market is shifting toward broader, system-wide clinical AI, but the strongest direct evidence for that shift is still company-authored rather than independently audited. SM024, SM025
CM046 Cancer-burden data support oncology as a plausible adjacency for Aidoc, but not yet a proven core revenue driver from the fetched evidence. SM013, SM014, SM015
CM047 Because Aidoc does not publicly disclose pricing, TAM/SAM/SOM work in this chapter must rely on health-spend proxies, hospital counts, and vendor deployment evidence rather than unit economics. SM011, SM012, SM024
CM048 The most important adverse evidence for this market today is the combination of reimbursement bottlenecks and the continued need for radiologist oversight when AI outputs are discordant. SM019, SM021, SM022
CP001 Aidoc says its radiology product integrates with EHR, PACS, scheduling, and reporting systems. SP001
CP002 Aidoc describes aiOS as a platform that runs, orchestrates, and governs clinical AI across a health system. SP002
CP003 Aidoc says aiOS can run multiple algorithms on a single scan and surface AI insights in the patient-record context. SP002
CP004 Aidoc's care-coordination product is designed to activate cross-specialty teams beyond image triage alone. SP003
CP005 Aidoc's 2026 official release says the company is deployed across nearly 2,000 hospitals and analyzes more than 60 million patient cases annually. SP004
CP006 Aidoc's 2025 financing materials say 69% of its customers already run non-Aidoc models on aiOS. SP025
CP007 AWS Marketplace lists Aidoc as available only by private offer. SP026
CP008 Viz.ai markets itself as an AI-powered care coordination platform with more than 50 FDA-cleared algorithms. SP005
CP009 Viz.ai said in its Series D announcement that the number of hospitals using the Viz Platform had surpassed 1,000. SP006
CP010 Viz.ai describes its differentiation as combining image-triggered disease detection with coordinated downstream action across clinical teams. SP005, SP006
CP011 Enlitic positions its offering around imaging-data standardization and workflow tools for radiologists, PACS administrators, IT professionals, and hospital administrators. SP007
CP012 Enlitic's public messaging emphasizes workflow infrastructure more than enterprise care coordination. SP007
CP013 Harrison.ai says Annalise's chest X-ray solution can identify up to 124 findings. SP008
CP014 Harrison.ai said in 2024 that Annalise had been selected by six NHS imaging networks covering 64 NHS trusts and 2.8 million chest X-rays annually. SP008
CP015 Nanox combines AI and software with a broader imaging-network and hardware commercialization model. SP009, SP010
CP016 Nanox reported only $0.5 million of Q4 2025 revenue from AI and Software solutions. SP009
CP017 PathAI describes AISight as a cloud-native digital pathology image-management and workflow platform used by laboratories and research centers. SP011
CP018 Roche agreed in 2026 to acquire PathAI for $750 million upfront plus up to $300 million in milestone payments. SP012
CP019 Paige positions itself around diagnostic AI, biomarker AI, pathology foundation models, and a pathology copilot rather than radiology triage. SP013
CP020 Tempus acquired Paige for $81.25 million and highlighted Paige's nearly 7 million digitized slide images. SP014
CP021 Microsoft markets Precision Imaging Network as a workflow layer that can add new AI models after a single contract, BAA, MSA, and security review. SP015, SP016
CP022 Nuance's partner ecosystem is designed to distribute imaging insights within existing workflows to radiologists, care providers, health plans, self-insured employers, and life-science stakeholders. SP016
CP023 Sectra markets a SaaS enterprise imaging platform that includes radiology, pathology, cardiology, and an AI marketplace. SP017
CP024 Sectra explicitly says customers can find, purchase, and deploy vetted AI applications from trusted partners inside one application. SP017
CP025 AGFA markets enterprise imaging as a unified platform with embedded AI rather than as a stand-alone clinical-AI point solution. SP018
CP026 Fujifilm's Synapse AI Orchestrator uses an open rules engine to bring more than 50 validated algorithms directly into PACS workflows. SP019
CP027 Fujifilm says University Radiology Group rolled Synapse AI Orchestrator across 37 facilities. SP019
CP028 GE says Edison offers more than 100 AI developer services and integrates with ACR AI-LAB to support compliant algorithm deployment. SP020
CP029 GE said its planned Intelerad acquisition would add cloud PACS, workflow orchestration, and a SaaS recurring-revenue model to its imaging business. SP021
CP030 GE said more than 1,500 healthcare organizations rely on Intelerad products. SP021
CP031 ACR's ARCH-AI program frames safe imaging-AI deployment as a governance and quality-assurance discipline. SP022
CP032 Radiology Business reported that imaging-AI vendors are shifting from single-use-case algorithms toward workflow orchestration, analytics, and enterprise value. SP023
CP033 Radiology Business reported that many imaging-AI products still struggle to show attractive ROI unless benefits are measured across the hospital network rather than only inside radiology. SP023
CP034 Aidoc's May 2025 FDA 510(k) clearance for BriefCase-Triage covers radiological computer-aided triage and notification software for aortic dissection. SP024
CP035 Aidoc's retained public materials show breadth across radiology triage, enterprise orchestration, and care coordination. SP001, SP002, SP003
CP036 Viz.ai remains the closest direct competitor because its public positioning still combines acute detection with coordinated downstream action across the care team. SP005, SP006
CP037 Enlitic, Annalise, and Nanox compete more on imaging workflow, finding breadth, and installed infrastructure than on Aidoc-style cross-specialty care coordination. SP007, SP008, SP009
CP038 PathAI and Paige are adjacent competitors because pathology foundation-model platforms can win enterprise AI budget and governance attention even without radiology-first workflows. SP011, SP013, SP018, SP020
CP039 Nuance/Microsoft, Sectra, Fujifilm, AGFA, GE, and Intelerad control workflow layers hospitals already buy, giving them durable distribution advantages over standalone AI vendors. SP015, SP017, SP018, SP019, SP021
CP040 Aidoc's moat depends more on workflow integration, governance, and enterprise AI consolidation than on any single algorithm lead. SP002, SP022, SP023, SP025
CP041 Because Aidoc and Fujifilm both market open orchestration for third-party models, multi-vendor hosting is becoming table stakes rather than a unique feature. SP002, SP019, SP025
CP042 Existing PACS, worklists, and governance processes remain a real substitute for buying a new enterprise clinical-AI operating layer. SP018, SP022, SP023
CI001 Aidoc's public product materials position the company as enterprise clinical AI embedded in hospital workflows rather than as self-serve software. SI008, SI010
CI002 Aidoc describes aiOS as a platform that runs and governs clinical AI across a health system. SI007
CI003 Aidoc's care-coordination offering extends the commercial story beyond radiology triage into downstream patient management and follow-up. SI009
CI004 AWS Marketplace lists Aidoc as available via private offer only. SI010
CI005 Aidoc's official product pages do not publish list prices or per-study rates. SI007, SI008, SI009, SI010
CI006 ITQlick estimates a first-year Aidoc deployment cost of $50,000 to $150,000-plus and says pricing starts around $50,000 per installation. SI011
CI007 Aidoc's public commercialization therefore looks enterprise and site-license oriented rather than transparently transactional or self-serve. SI008, SI009, SI010
CI008 Aidoc says its radiology product integrates with PACS, EHR, scheduling, and reporting systems. SI008
CI009 Aidoc said in 2025 that 69% of its customers were already running non-Aidoc models on aiOS. SI002, SI007
CI010 Aidoc's 2023 official funding release said a $110 million Series D brought total funding to $250 million. SI001
CI011 Aidoc's 2025 growth-financing release said $150 million of financing plus a $40 million revolving credit facility brought total funding to $370 million. SI002
CI012 Aidoc's 2026 official and Goldman releases both announced a $150 million Series E. SI003, SI004
CI013 Aidoc's 2026 official and Goldman releases say total funding is now over $500 million. SI003, SI004
CI014 Aidoc's 2023, 2025, and 2026 lifetime-funding figures do not reconcile cleanly with one another. SI001, SI002, SI003
CI015 Calcalist reported that Aidoc's 2025 financing was raised at a valuation higher than its previous round. SI005
CI016 Globes reported that Aidoc would not disclose its revenue or the valuation at which the 2025 round was raised. SI006
CI017 Post-2024 public evidence does not confirm a numeric Aidoc valuation or cleanly confirm current unicorn status. SI004, SI005, SI006
CI018 Aidoc said in 2025 that it supports care for more than 45 million patients annually across 150 or more health systems. SI002
CI019 Aidoc said in 2026 that it analyzes more than 60 million patient cases annually and is deployed across nearly 2,000 hospitals. SI003
CI020 AWS Marketplace says Aidoc is used by over 1,600 hospitals worldwide. SI010
CI021 Aidoc's public scale disclosures mix health-system counts, hospital counts, and patient volumes, which limits direct revenue modeling from the disclosed data. SI002, SI003, SI010
CI022 Goldman said health systems using Aidoc report shorter lengths of stay and measurable financial returns. SI004
CI023 Aidoc's length-of-stay infographic frames the product's commercial value around faster diagnosis-to-treatment-to-discharge and hospital efficiency. SI012, SI023
CI024 Aidoc's K251406 filing shows the company received FDA clearance for BriefCase-Triage in May 2025. SI013
CI025 Public evidence shows Aidoc remains a regulated triage-software seller even as it shifts toward foundation-model messaging. SI003, SI013
CI026 Radiology Business reported that global medical imaging AI generated about $749 million in revenue in 2024. SI014
CI027 Radiology Business reported that only a small number of imaging-AI applications currently have reimbursement through Category 1 CPT codes. SI014
CI028 Radiology Business reported that imaging-AI ROI often does not make sense unless benefits extend across the healthcare system. SI014
CI029 ACR's ARCH-AI program indicates that enterprise imaging AI requires formal governance and quality assurance. SI015
CI030 GE said Intelerad would contribute about $270 million of revenue, roughly 90% recurring revenue, and more than 30% adjusted EBITDA margin in GE's first full year of ownership. SI016
CI031 Intelerad's public benchmark shows that mature workflow-layer imaging software can have attractive recurring-revenue economics. SI016
CI032 Nanox reported only $0.5 million of Q4 2025 revenue from AI and Software solutions and said it expected to need additional financing to implement its business plan. SI017
CI033 Nanox illustrates that public imaging-AI economics can remain low-scale and loss-making even with commercial deployment activity. SI017
CI034 Tempus acquired Paige for $81.25 million. SI018
CI035 Roche agreed to buy PathAI for $750 million upfront plus up to $300 million of milestones. SI019
CI036 Viz.ai's 2022 Series D valued the company at $1.2 billion. SI020
CI037 Public comp transactions across Paige, PathAI, and Viz.ai provide only a wide and noisy valuation-input band for Aidoc. SI018, SI019, SI020
CI038 Microsoft markets procurement simplicity—a single contract, BAA, and security review—as a core selling point for imaging-AI deployment. SI021, SI022
CI039 Aidoc therefore must win on deployment friction and workflow fit as well as on algorithm quality. SI010, SI021, SI022, SI025, SI026
CI040 Aidoc's disclosed capital history reduces near-term distress risk after the 2026 round, but capital adequacy cannot be underwritten without cash, burn, and debt data. SI001, SI002, SI003, SI006
CI041 Public sources do not disclose Aidoc's ARR, absolute revenue, gross margin, CAC, payback, NRR, cash balance, or debt schedule. SI006, SI007, SI008, SI009
CI042 Calcalist quoted Aidoc's CTO saying revenue from the new technology already surpasses the company's recent years and could exceed the previous nine years within a year, while still not disclosing the absolute revenue number. SI005
CI043 Because pricing is opaque and public scale disclosures mix incompatible units, a supportable ARR estimate is not possible from public evidence alone. SI005, SI006, SI010, SI011
CI044 The 2025 financing included a $40 million revolving credit facility, so Aidoc's capital stack is not purely equity. SI002
CI045 Aidoc raised again less than a year after the 2025 growth financing, implying continued capital appetite while scaling CARE and aiOS. SI002, SI003, SI024
CI046 Aidoc's public product and marketplace signals support a SaaS / enterprise-license commercialization model, while public evidence does not support per-study pricing as the dominant disclosed mechanism. SI007, SI008, SI009, SI010
CI047 Aidoc's 2026 newsroom index shows post-Series-E announcements spanning new deployments, European product expansion, and leadership additions, indicating continued operating momentum after financing. SI024, SI027
CI048 Sol Radiology deployed a suite of FDA-cleared Aidoc AI solutions across hospitals, outpatient imaging centers, and urgent care facilities in Southern California. SI028
CI049 Aidoc's Europe triage release says aiOS now offers 35 AI solutions approved for the European market, including 28 self-developed algorithms and 7 partner-developed solutions. SI030
CI050 Aidoc's PR Newswire announcement for new CMO Jesse Ehrenfeld said aiOS had surpassed 100 million patient cases analyzed and was deployed in more than 1,600 hospitals globally. SI031
CI051 Isala Hospital's AI-powered pulmonary embolism response workflow uses Aidoc across more than 70 clinicians and is linked to a 14-hospital Dutch trial context, indicating workflow-level utility beyond the U.S. market. SI029
CE001 Aidoc describes aiOS as the platform that runs, orchestrates and governs clinical AI across a health system. SE001, SE017
CE002 Aidoc says aiOS uses textual data, scan metadata and pixel analysis to decide which studies receive AI processing. SE001
CE003 aiOS is marketed as a layer that can run multiple algorithms on a single scan and present insights in patient context. SE001
CE004 Aidoc publicly states that aiOS integrates with PACS, EHR, mobile and care-team workflows. SE001, SE006, SE007
CE005 The aiOS page says the platform provides validation, drift detection, override tracking and analytics for governance. SE001
CE006 Aidoc markets aiOS as a one-integration platform that lets health systems scale AI without re-architecting infrastructure. SE001, SE003
CE007 Aidoc describes CARE as a clinical-grade foundation model trained on real-world multimodal data. SE002, SE023
CE008 The CARE page says the model can be pretrained on imaging, text, EHR, labs and vitals rather than on a single data type. SE002
CE009 Healthcare IT Today reported Aidoc’s claim that CARE enables development of new indications up to 20 times faster. SE023
CE010 Aidoc explicitly pairs CARE and aiOS as the combined foundation-model-plus-platform architecture for scalable clinical intelligence. SE001, SE002
CE011 Aidoc’s January 2026 release said the new body CT workflow combines 11 newly cleared indications with three previously cleared indications in one workflow. SE003, SE021
CE012 Aidoc’s January 2026 release reported mean sensitivity of 97% and mean specificity of 98% across the 11 newly cleared indications, plus roughly an order-of-magnitude fewer false alerts versus best-in-class single-condition tools. SE003, SE021
CE013 FDA summary K252970 lists diverticulitis, abdominal-pelvic abscess, appendicitis, intestinal ischemia or pneumatosis, obstructive renal stone, small and large bowel obstruction, spleen injury, liver injury, kidney injury and pelvic fracture among the newly cleared findings. SE004
CE014 FDA summary K252970 defines CARE Multi-triage CT Body as triage and notification software for workflow prioritization, not diagnostic software. SE004, SE009
CE015 FDA summary K252970 says the cleared software runs on a Linux-based server in a cloud environment and works with DICOM CT images, PACS and radiology workstations. SE004
CE016 Aidoc’s neuro page markets a Full Brain solution spanning vessel occlusion, CT perfusion, brain aneurysm, intracranial hemorrhage, C-spine fracture and vertebral compression fracture workflows. SE005
CE017 Aidoc’s VTE page positions the product as a full VTE care continuum covering PE, incidental PE, DVT, RV/LV coordination and IVC-filter follow-up rather than only PE triage. SE006
CE018 The VTE page says aiOS delivers PE, iPE, DVT and related follow-up insights into PACS, EHR and mobile workflows with PERT activation. SE006
CE019 Aidoc’s aortic page combines acute aortic dissection triage, aneurysm follow-up and patient-management workflows in one product family. SE007
CE020 The aortic page says care coordination includes mobile alerts, image review, deep EHR connectivity, cross-department chat and care activation. SE007
CE021 Aidoc’s quality page says the QMS is certified to MDSAP and ISO 13485 and compliant with FDA QSR, EU MDR, ISO 14971 and IEC 62304. SE009
CE022 Aidoc’s security page says the platform aligns to NIST CSF and uses AWS/Azure plus EDR, encryption, SIEM and CSPM for data protection. SE008
CE023 Aidoc’s AWS page describes the product as a HIPAA-compliant, scalable, secure enterprise healthcare AI deployment on AWS. SE011
CE024 BRIDGE is publicly described as a structured framework developed with NVIDIA, health systems and industry partners to move AI from pilots to scalable responsible deployment. SE012, SE018
CE025 Aidoc’s careers page lists open roles across AI algorithms, ML platform, DevOps/cloud, backend, technical operations and infrastructure product management. SE013
CE026 Aidoc’s April 2026 releases say the company analyzes more than 60 million patient cases annually, has processed more than 110 million total cases, and is deployed across nearly 2,000 hospitals. SE014, SE015, SE016, SE017
CE027 Fierce Healthcare and Healthcare IT Today reported that aiOS hosts both Aidoc and third-party models, with 69% of customers running non-Aidoc models on the platform. SE018, SE023
CE028 Aidoc’s 2026 public roadmap says CARE will expand across CT and X-ray workflows and add automated draft report creation in the near term. SE003, SE014, SE015
CE029 Aidoc’s official pages repeatedly market the “largest portfolio of FDA-cleared algorithms” without publishing a current reconciled numeric total on site. SE001, SE023
CE030 A third-party FDA compilation site counts 34 Aidoc 510(k) clearances by 2026, while Fierce cited 18 FDA-cleared algorithms in 2025, so public portfolio totals remain metric-dependent and unreconciled by Aidoc. SE022, SE018
CE031 Hartford HealthCare’s official partnership page and HIT Consultant both say the initial aiOS go-live took three weeks and included 17 FDA-cleared algorithms across millions of patient exams. SE024, SE027
CE032 Mercy’s public deployment story says aiOS was live across all 50 Mercy facilities by February 2025 with over a dozen use cases running simultaneously. SE025
CE033 Mercy’s public deployment story says the health system analyzed 2.4 million images in the prior year, flagged 249,000 studies and reported a 90% reduction in outpatient time-to-diagnosis. SE025
CE034 Asklepios said Aidoc was active at 28 hospitals by early 2026 and analyzing approximately 35,000 CT and X-ray images monthly. SE026
CE035 Asklepios described its rollout as a secure cloud-based approach with an on-prem aiOS platform to satisfy GDPR and existing radiology-system integration requirements. SE026
CE036 Public materials still do not provide a detailed reference architecture for third-party model onboarding, customer-by-customer cloud/on-prem mix, or independent multi-site benchmark audits for CARE. SE001, SE012, SE026
CE037 Diagnostic Imaging summarized the January 2026 release as a CT-based AI triage platform with 14 total cleared indications. SE021, SE003
CE038 FDA summary K252970 says the CARE Multi-triage CT Body device was compared against Aidoc’s Aortic Dissection predicate and that both devices contain modules fine-tuned from a locked foundation model. SE004
CU001 Aidoc’s April 2026 official and PRNewswire releases say the company is deployed across nearly 2,000 hospitals and analyzes more than 60 million cases annually. SU019, SU020
CU002 Independent 2025 articles repeated Aidoc’s claim of supporting care for more than 45 million patients annually across 150-plus health systems, with a goal of reaching 100 million patients in three years. SU012, SU013, SU015
CU003 Healthcare IT Today, MedCity News and HLTH all said Hartford, Mercy, Sutter and WellSpan joined Aidoc’s 2025 financing round as strategic health-system investors. SU012, SU014, SU015
CU004 Independent 2025 coverage also named Mount Sinai, Yale New Haven, Northwell, University of Miami and Temple among Aidoc partners or users. SU012, SU013, SU014
CU005 Aidoc’s Hartford announcement and HIT Consultant both said the initial go-live took three weeks and deployed 17 FDA-cleared algorithms across millions of patient exams. SU002, SU003
CU006 The Hartford deployment spans radiology, cardiology, vascular, neurology and emergency workflows across a 500-plus-location Connecticut care system. SU002, SU003
CU007 Hartford said full enterprise implementation of Aidoc was planned over the following 12 months after launch. SU002, SU003
CU008 Mercy’s public customer story says Aidoc had already analyzed over one million images across the health system and flagged more than 120,000 critical findings in real time. SU005
CU009 Mercy’s public deployment materials say aiOS was live across all 50 Mercy facilities by February 2025 with more than a dozen use cases running simultaneously. SU006, SU024, SU025
CU010 Mercy’s deployment story says the prior year produced 2.4 million images analyzed, 249,000 flagged studies and a 90 percent reduction in outpatient time-to-diagnosis. SU006
CU011 Asklepios said Aidoc was active at 28 hospital locations and analyzing about 35,000 CT and X-ray images monthly by early 2026. SU007
CU012 Sutter Health’s official announcement says the partnership will embed aiOS across Sutter’s care system serving more than 3.5 million Californians and make Sutter Aidoc’s West Coast hub. SU008
CU013 Yale New Haven case-study sources say AI-triggered PERT activation increased appropriate advanced-therapy use by about 40 percent and surfaced roughly 70 percent of potential activations that otherwise would have been missed. SU009, SU021
CU014 Aidoc’s Renown and Carson Tahoe blog says door-in-door-out time fell by 32 minutes, LVO transfer time by roughly 30 percent and door-to-needle time by roughly 16 percent after around six months. SU010
CU015 Temple Health’s CEO said Aidoc stood out on PACS/EHR integration, breadth across FDA-approved solutions and operational fit after about a year and a half of use. SU011
CU016 Aidoc’s VTE page cites a 40 percent increase in PERT consultations at Yale New Haven Health. SU021
CU017 Aidoc’s VTE page cites Cedars-Sinai metrics of a 7-hour (41 percent) reduction in time-to-treatment and a 26 percent reduction in length of stay for pulmonary embolism workflows. SU021
CU018 Aidoc’s aiOS page cites Advocate Health with 69 hospitals, more than 8 million annual imaging studies, a 22-site rollout and a projected 63,000 patients per year benefiting from faster triage. SU018
CU019 Healthcare IT Today quotes WellSpan’s CEO saying Aidoc helped radiologists analyze more than 200,000 cases in one year and significantly reduce diagnosis delays. SU012
CU020 FeaturedCustomers says Aidoc has 28 reviews or testimonials, 1 case study, 22 customer videos and 1,448 reference ratings. SU016
CU021 FeaturedCustomers is an aggregated marketplace proof source, so its counts show public references exist but do not prove representativeness across Aidoc’s installed base. SU016
CU022 ITQlick’s 2026 review says Aidoc deployments may cost roughly 50,000 to 150,000 dollars for small practices and exceed 500,000 dollars for large enterprises, but the site is not an audited primary pricing source. SU017
CU023 ITQlick also flags limited pricing transparency and integration burden as barriers for smaller clinics. SU017
CU024 Fetched public sources for this chapter do not disclose NRR, GRR, renewal-rate or churn metrics for Aidoc. SU012, SU013, SU014, SU016
CU025 Fetched public sources for this chapter do not disclose top-customer concentration, revenue-share concentration or named account ARR. SU012, SU013, SU014
CU026 Public named-customer proof is concentrated in Hartford, Mercy, Asklepios, Sutter, Yale, Renown or Carson Tahoe and Temple rather than a broad disclosed reference base. SU001, SU002, SU005, SU007, SU008, SU009, SU010, SU011
CU027 Cedars-Sinai appears in fetched sources as a study site for VTE workflow outcomes, not as a directly documented enterprise-wide Aidoc rollout. SU021
CU028 Aidoc’s 150-plus-health-system and 45-million-patient figures are repeated across multiple independent 2025 sources but remain company-reported scale metrics rather than independently audited census counts. SU012, SU013, SU015
CU029 Public customer scale metrics vary by denominator—150-plus health systems, 45-million patients, nearly 2,000 hospitals and 60-million cases—so those figures should not be treated as interchangeable customer counts. SU019, SU020, SU012, SU013
CU030 Sutter’s official announcement frames the relationship as a multi-year infrastructure and co-development partnership rather than a one-off algorithm purchase. SU008
CU031 Mercy’s deployment story says platform standardization is what allowed the system to scale beyond a few isolated AI tools. SU006
CU032 Temple’s procurement discussion shows that ROI scrutiny is a gating factor in Aidoc’s sales motion rather than an afterthought. SU011
CU033 Renown and Carson Tahoe explicitly label their published impact figures as internal site data. SU010
CU034 Asklepios positions Aidoc as part of its broader Health Data Hub and CDSS strategy and highlights value for smaller regional sites as well as large hospitals. SU007
CU035 Hartford’s trust-focused blog and launch materials emphasize alignment and anecdotal early wins, but they do not publish ROI or retention cohorts. SU004, SU002
CU036 Mercy executives’ claim that every Mercy patient benefits from Aidoc is directional executive testimony, not a measured utilization rate across all encounters. SU006
CU037 Healthcare IT Today’s report that 69 percent of customers run non-Aidoc models on aiOS implies many buyers treat Aidoc as an orchestration layer for multi-vendor AI estates. SU012
CU038 Publicly fetched sources directly confirm Hartford, Mercy, Sutter, Asklepios, Yale, Renown or Carson Tahoe, Temple, WellSpan and Advocate, but do not directly confirm NYU Langone, Mayo Clinic or the University of Rochester Medical Center; that is a proof gap, not proof of absence. SU002, SU005, SU007, SU008, SU009, SU010, SU011, SU012, SU018
CR001 Aidoc says its quality system is certified to MDSAP, aligned with FDA quality-system rules, ISO 13485, EU MDR, ISO 14971, and IEC 62304. SR001
CR002 Aidoc announced in January 2026 that the FDA cleared a comprehensive AI triage solution combining 11 newly cleared indications with three existing ones. SR008
CR003 FDA’s January 2025 draft guidance requires AI-enabled device software sponsors to document lifecycle management and marketing-submission evidence more explicitly. SR026
CR004 FDA’s August 2025 PCCP guidance shows that post-market model updates for AI-enabled devices increasingly need predefined change-governance plans. SR027, SR028
CR005 Aidoc’s BRIDGE framework publicly frames trust, compliance, and workflow integration as prerequisites for scaled clinical AI adoption. SR005
CR006 Aidoc’s public sources did not disclose a 2026 CE-MDR update for the new comprehensive foundation-model product, leaving European regulatory status partially opaque. SR001, SR008
CR007 Clinical AI triage that changes case prioritization and team activation creates liability exposure if a missed or delayed finding contributes to patient harm. SR020, SR029
CR008 A peer-reviewed study of an AI-augmented radiology worklist triage system reported lower length of stay for intracranial hemorrhage and pulmonary embolism after adoption. SR020
CR009 Aidoc customer and health-system releases repeatedly market faster identification of pulmonary embolism, stroke, and emergency findings as a core value proposition. SR011, SR014, SR015, SR016, SR017, SR018, SR019
CR010 Because Aidoc’s products now span multiple countries, hospitals, and care settings, site heterogeneity and domain shift remain material performance risks. SR011, SR012, SR013, SR039, SR040
CR011 A 2025 review in Diagnostic and Interventional Radiology states that medical-imaging AI can be compromised by several forms of bias that may adversely affect patient outcomes. SR039
CR012 A 2021 radiology AI review argues that failure to generalize across new data distributions is one of the main obstacles to safe clinical deployment. SR040
CR013 Aidoc’s public security page highlights NIST Cybersecurity Framework adoption and AWS hosting but does not publicly enumerate SOC 2 or HITRUST attestations. SR002, SR006
CR014 Aidoc’s U.S. data-transfer notice says the company certified to the EU-U.S., UK extension, and Swiss-U.S. Data Privacy Frameworks. SR003
CR015 HHS OCR’s April 2026 ransomware settlement announcement shows HIPAA security enforcement remains active and focused on cyber controls after breaches affecting more than 427,000 individuals. SR031, SR032
CR016 HHS’ resolution-agreement page confirms that OCR settlements can require multi-year monitoring and corrective-action obligations on covered entities and business associates. SR030, SR032
CR017 HIPAA violation cases can be resolved with financial penalties when OCR identifies serious or systemic noncompliance. SR035, SR030
CR018 Aidoc’s partner page says aiOS is vendor-agnostic and connects to PACS, VNA, worklists, EHR, scheduling, and communication interfaces. SR004
CR019 Aidoc’s partner page says the company is available within Epic App Orchard and can integrate with Epic Radiant for acuity-based feedback. SR004
CR020 Microsoft’s Epic-focused healthcare page positions Dragon and Azure AI as embedded workflow tools for clinicians, increasing incumbent-platform competition. SR037
CR021 Oracle Health markets a clinical AI agent that drafts documentation, automates coding and scheduling, and coordinates workflows across clinical and administrative roles. SR038
CR022 Epic’s official AI page describes AI as embedded throughout its software for patients, clinicians, and operations. SR036
CR023 Aidoc and AWS announced a multiyear strategic collaboration with significant AWS investment focused on optimizing the CARE foundation model. SR022, SR023, SR024
CR024 Radiology Business reported that the AWS investment terms were undisclosed even as Aidoc described the arrangement as significant and multiyear. SR023, SR022
CR025 Aidoc’s AWS partnership and AWS-hosted platform reduce infrastructure buildout burden but increase dependence on one cloud and strategic partner. SR006, SR022, SR023
CR026 Advocate Health said its Aidoc agreement followed a successful pilot and could benefit nearly 63,000 patients annually. SR018
CR027 Novant Health said it deployed seven FDA-cleared Aidoc solutions during emergency-department strain, showing hospitals may buy AI triage when workflow ROI is immediate. SR019
CR028 Asklepios said its group-wide rollout covers 28 hospitals and about 35,000 CT and X-ray images monthly, which demonstrates traction but also raises implementation-support burden. SR012
CR029 Sol Radiology’s 2026 deployment shows Aidoc is extending beyond flagship hospitals into physician-led radiology organizations and outpatient-adjacent settings. SR013
CR030 Radiology Business reported that reimbursement still lags radiology AI adoption because most tools use Category III rather than payment-linked Category I CPT codes. SR034, SR033
CR031 The ACR describes AI reimbursement in radiology as an evolving practice rather than a settled payment regime. SR033
CR032 Aidoc’s January 2026 CMO announcement said aiOS had surpassed 100 million patient cases analyzed. SR009
CR033 Aidoc’s leadership page lists a broader executive bench including CEO, CTO, chief architect, president and chief commercial officer, chief product officer, physician executive, chief R&D and AI officer, chief people officer, and chief medical officer. SR009, SR004
CR034 Jesse Ehrenfeld joined Aidoc as chief medical officer after serving as AMA president, which strengthens clinical-policy credibility. SR009
CR035 Aidoc raised $150 million in Series E in April 2026 and said total funding exceeded $500 million. SR010
CR036 Aidoc said the Series E came less than a year after a growth round led by General Catalyst and Square Peg, implying a faster financing cadence and higher investor expectations. SR010
CR037 Large late-stage financings can mitigate near-term runway risk while raising the threshold for operating execution and next-round proof. SR010, SR023
CR038 Public sources reviewed did not identify a specific Aidoc data breach, recall, or public OCR enforcement action as of 2026-05-20. SR030, SR031, SR032
CR039 The absence of a public adverse-event record is not equivalent to audited evidence that such events have not occurred. SR030, SR031, SR035
CR040 Risk to the investment thesis rises materially if Aidoc loses privileged interoperability status in Epic or other core workflow systems. SR004, SR036, SR037
CR041 Risk to the investment thesis rises materially if reimbursement remains pilot-like and customers cannot anchor Aidoc spend to durable budget or payment lines. SR033, SR034, SR018
CR042 Risk to the investment thesis rises materially if foundation-model updates outpace the company’s ability to maintain FDA-compliant change control and safety evidence. SR026, SR027, SR028, SR008
CR043 Aidoc’s combination of QMS certifications, enterprise integrations, customer deployments, and new clinical leadership provides real mitigation, but not enough to eliminate regulatory, liability, and procurement risk. SR001, SR004, SR009, SR018
CR044 The most material residual risks are FDA and model-governance burden, liability from missed findings, payment opacity, platform dependence, and execution against a rapidly rising financing bar. SR026, SR027, SR030, SR034, SR010
CV001 Aidoc’s last fully public round with a clearly disclosed amount was a $110 million Series D announced on June 16, 2022, bringing total funding to $250 million. SV001, SV002
CV002 The 2022 primary round announcement did not disclose an exact post-money valuation. SV001, SV002
CV003 Aidoc announced a $150 million Series E in April 2026 and said total funding exceeded $500 million. SV003, SV010
CV004 Aidoc’s 2026 Series E was led by Growth Equity at Goldman Sachs Alternatives with participation from General Catalyst, SoftBank Vision Fund 2, and NVentures. SV003
CV005 Aidoc’s January 2026 CMO announcement said aiOS had surpassed 100 million patient cases analyzed. SV004
CV006 Aidoc’s January 2026 FDA clearance strengthened product proof but did not disclose monetization or revenue. SV005
CV007 Aidoc’s 2025-2026 customer evidence includes Advocate Health and Asklepios enterprise rollouts rather than isolated pilots. SV006, SV007
CV008 Aidoc’s AWS collaboration included significant multiyear investment directed at the CARE foundation model. SV008, SV031
CV009 CTech reported in July 2025 that Aidoc raised $150 million at a valuation higher than its previous round but did not publish an exact post-money number. SV009
CV010 Fierce Healthcare’s April 2026 funding coverage also omitted an exact valuation while emphasizing expansion and clinical-AI momentum. SV010
CV011 The combination of undisclosed post-2024 financing terms and conflicting public financing chronology means Aidoc’s current private valuation cannot be independently confirmed from public evidence. SV003, SV009, SV010
CV012 Because no exact post-2024-05-20 valuation is publicly disclosed, Aidoc’s current unicorn status should be treated as unconfirmed rather than asserted as fact. SV003, SV009, SV010
CV013 Viz.ai raised $100 million at a $1.2 billion valuation in 2022 after surpassing 1,000 hospitals using its platform. SV011
CV014 Abridge’s 2025 Series E raised $300 million and said the company was partnering with more than 150 enterprise health systems. SV012
CV015 Roche agreed in 2026 to acquire PathAI for $750 million upfront plus up to $300 million of additional milestones. SV013
CV016 PathAI’s last major disclosed primary financing before the Roche deal was a $165 million Series C in 2021. SV014
CV017 Tempus acquired Paige in 2025 for $81.25 million, paid predominantly in stock, plus assumption of Paige’s remaining Azure commitment. SV015
CV018 Tempus had a market capitalization of about $8.17 billion in May 2026 on the cited CompaniesMarketCap snapshot. SV016
CV019 Tempus SEC companyfacts show roughly $904.6 million of revenue at September 30, 2025. SV021
CV020 Tempus therefore traded at roughly 9.0x market-cap-to-revenue on the cited anchors. SV016, SV021
CV021 RadNet had a market capitalization of about $4.19 billion in May 2026 on the cited CompaniesMarketCap snapshot. SV017
CV022 RadNet SEC companyfacts show roughly $1.492 billion of revenue at September 30, 2025. SV022
CV023 RadNet therefore traded near 2.8x market-cap-to-revenue on the cited anchors. SV017, SV022
CV024 Phreesia had a market capitalization of about $0.56 billion in May 2026 on the cited CompaniesMarketCap snapshot. SV018
CV025 Phreesia SEC companyfacts show roughly $353.5 million of revenue at October 31, 2025. SV023
CV026 Phreesia therefore traded near 1.6x market-cap-to-revenue on the cited anchors. SV018, SV023
CV027 Health Catalyst had a market capitalization of about $93.84 million in May 2026 on the cited CompaniesMarketCap snapshot. SV019
CV028 Health Catalyst SEC companyfacts show roughly $236.5 million of revenue at September 30, 2025. SV024
CV029 Health Catalyst therefore traded near 0.4x market-cap-to-revenue on the cited anchors. SV019, SV024
CV030 GE HealthCare had a market capitalization of about $28.01 billion in May 2026 on the cited CompaniesMarketCap snapshot. SV020
CV031 GE HealthCare SEC companyfacts show roughly $14.927 billion of revenue at September 30, 2025. SV025
CV032 GE HealthCare therefore traded near 1.9x market-cap-to-revenue on the cited anchors. SV020, SV025
CV033 The public comparator set spans roughly 0.4x to 9.0x on current market-cap-to-revenue anchors, with Tempus as the premium AI outlier and mainstream healthcare IT or imaging nearer 1x to 3x. SV016, SV017, SV018, SV019, SV020, SV021, SV022, SV023, SV024, SV025
CV034 Rock Health’s 2025 year-end report described a two-speed digital-health market rather than a broad-based reopening. SV026
CV035 Rock Health’s Q1 2026 report said digital-health capital was still concentrating rather than normalizing. SV027
CV036 Cooley’s Q4 2025 venture report showed late-stage venture terms improved from the trough but remained selective rather than easy. SV028
CV037 Radiology Business and the ACR continue to frame reimbursement as a gating factor for radiology-AI monetization. SV029, SV030
CV038 Aidoc’s strongest public valuation supports are product traction, enterprise adoption, regulatory progress, and strategic investors, not disclosed revenue or disclosed price. SV003, SV005, SV006, SV007, SV008
CV039 Without disclosed ARR, gross margin, NRR, or exact round terms, a precise present-day Aidoc entry multiple cannot be underwritten. SV003, SV009, SV010
CV040 A bull case requires Aidoc to convert current deployment momentum into several hundred million dollars of recurring revenue so that a Tempus-like or category-premium multiple becomes relevant. SV016, SV021, SV003, SV007, SV008
CV041 A base case assumes Aidoc eventually prices closer to premium public health-AI and imaging multiples than to opaque private scarcity premiums. SV017, SV018, SV020, SV021, SV022, SV023, SV025
CV042 A bear case assumes reimbursement drag, procurement friction, or EHR-native competition compress valuation toward 1x to 3x public healthcare-IT ranges. SV017, SV018, SV020, SV029, SV030
CV043 Recommendation is research-more until investors obtain the exact current price, revenue-quality data, and preference stack. SV003, SV009, SV010, SV026, SV028
CV044 Thesis strength comes from visible strategic and clinical traction: >$500 million raised, new FDA clearance, AWS backing, 100 million analyzed cases, and named enterprise deployments. SV003, SV004, SV005, SV006, SV007, SV008
CV045 Thesis-break triggers include any flat or down round versus undisclosed 2025-2026 pricing, failure to convert rollouts into paid expansion, or proof that reimbursement headwinds cap budgeted demand. SV026, SV027, SV028, SV029, SV030
CV046 Diligence still needs signed ARR by cohort, use-case gross margin, retention and expansion by deployment, exact round documents, and any valuation-support memo used in the latest financing. SV003, SV009, SV010
CV047 An illustrative bull exit case of roughly $4.5 billion to $7.0 billion requires Aidoc to reach about $450 million to $700 million of revenue at 8x to 10x premium multiples. SV016, SV021, SV026
CV048 An illustrative base exit case of roughly $1.8 billion to $3.4 billion requires Aidoc to reach about $300 million to $450 million of revenue at 5x to 7x multiples. SV017, SV018, SV020
CV049 An illustrative bear exit case of roughly $0.6 billion to $1.4 billion assumes revenue stalls near $150 million to $250 million and multiples compress to 2x to 4x. SV017, SV018, SV029, SV030
来源
编号出版方标题引文
SO001 Aidoc Meet Aidoc: Your Partner in Clinical AI We are a pioneering force in clinical AI focusing on aiding and empowering healthcare teams to optimize patient treatment, resulting in improved economic value and clinical outcomes.
SO002 Aidoc Elad Walach - Aidoc | Clinical AI Elad Walach is a co-founder and CEO of Aidoc.
SO003 Aidoc Guy Reiner - Aidoc | Clinical AI Guy Reiner is a co-founder, Chief Architect and GM of the Tel-Aviv Branch at Aidoc.
SO004 Aidoc Michael Braginsky - Aidoc | Clinical AI Michael Braginsky is a co-founder and CTO of Aidoc.
SO005 Aidoc aiOS™ | End-To-End Clinical AI Platform | Aidoc Aidoc’s aiOS™ is the clinical AI platform designed to run, orchestrate and govern clinical AI across your health system.
SO006 Aidoc Aidoc Foundation Model page CARE™ (Clinical AI Reasoning Engine) is Aidoc’s clinical-grade foundation model, trained on real-world, multimodal data.
SO007 Aidoc Care Coordination Solutions | Aidoc - Real-Time Impact Aidoc’s AI-powered Care Coordination Solutions connect care teams in real time, accelerating treatment decisions and improving outcomes across departments.
SO008 Aidoc Radiology AI Imaging | Aidoc – Faster, Smarter Care Aidoc’s advanced AI medical imaging helps radiologists streamline workflows, prioritize findings, activate care teams and facilitate patient follow-up.
SO009 Aidoc Aidoc raises $110M in Series D Round to Expand AI Care Platform Aidoc announced today a $110 million Series D round investment, bringing its total funding to $250 million.
SO010 PR Newswire Aidoc Secures $150M for CARE, its Healthcare Foundation Model, to Transform Clinical Decision-Making for 100 Million Patients The round was led by General Catalyst and Square Peg ... bringing the company's total funding to $370 million.
SO011 CTech by Calcalist Aidoc raises $150M with Nvidia backing as AI pushes faster diagnosis Aidoc was founded in 2016 by its CEO Elad Walach, CTO Michael Braginsky, and VP of R&D and CISO Guy Reiner.
SO012 Globes AI medical decision co Aidoc raises $150m The company does not disclose details about its revenue, nor the company valuation at which the funding was raised.
SO013 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses Aidoc has raised $150 million in Series E funding led by Growth Equity at Goldman Sachs Alternatives.
SO014 Goldman Sachs Asset Management Aidoc Raises $150M Series E Led by Goldman Sachs The round brings total funding to over $500 million, less than a year after a growth round led by General Catalyst and Square Peg.
SO015 Fierce Healthcare Aidoc banks $150M backed by Goldman Sachs to scale clinical AI foundation model Aidoc, founded in 2016, provides AI-powered tools in radiology, cardiology, neurovascular and vascular and plans to expand into oncology.
SO016 Aidoc Aidoc Secures Landmark FDA Clearance for Foundation Model AI This clearance applies to Aidoc’s Rib Fractures triage solution, a new version built on Aidoc’s CARE1™ Foundation Model.
SO017 Aidoc Aidoc Secures New FDA Clearance The solution brings 11 newly cleared indications and three previously cleared indications together into a single workflow.
SO018 Aidoc Asklepios Successfully Completes Comprehensive AI Rollout in Radiology: 28 Hospitals Using Aidoc to Support Patient Care The system is now active at 28 hospital locations and supports medical teams around the clock in analysing X-ray and CT images.
SO019 Aidoc Hartford HealthCare and Aidoc Partnership Hartford HealthCare has implemented Aidoc and its aiOS™ platform, featuring its 17 FDA-cleared algorithms, across millions of patient exams annually.
SO020 Coalition for Health AI / Aidoc Aidoc CAC-01 model card (raw XML) Aidoc's products are structured based on international quality, privacy, and security standards and frameworks, including ISO 13485, ISO 27001, ISO 27017, ISO 27018, ISO 27799, SOC 2 Type 2, Cyber Essentials, and C5.
SO021 U.S. Food and Drug Administration K231631 510(k) clearance letter for BriefCase-Quantification Re: K231631 Trade/Device Name: BriefCase-Quantification.
SO022 U.S. Food and Drug Administration K213721 510(k) clearance letter for BriefCase Re: K213721 Trade/Device Name: BriefCase.
SO023 U.S. Food and Drug Administration AI-Enabled Medical Devices The FDA AI-enabled devices roster includes multiple Aidoc clearances across 2022-2025 entries.
SO024 Aidoc Aidoc Clinical Compendium Aidoc's clinical compendium with 100+ Peer-reviewed publications or abstract/conference presentations are available at the following link.
SO025 PubMed Central Reimbursement in the age of generalist radiology artificial intelligence We argue that generalist radiology artificial intelligence challenges current healthcare reimbursement frameworks.
SO026 dotmed.com AI for pulmonary embolism detection shows high agreement with radiologists in real-world study Among confirmed pulmonary embolism cases, 15% were identified by radiologists but missed by the algorithm.
SM001 Aidoc Radiology AI Imaging | Aidoc – Faster, Smarter Care Aidoc’s advanced AI medical imaging helps radiologists streamline workflows, prioritize findings, activate care teams and facilitate patient follow-up.
SM002 Aidoc Care Coordination Solutions | Aidoc - Real-Time Impact Aidoc’s AI-powered Care Coordination Solutions connect care teams in real time, accelerating treatment decisions and improving outcomes across departments.
SM003 Aidoc Asklepios Successfully Completes Comprehensive AI Rollout in Radiology: 28 Hospitals Using Aidoc to Support Patient Care The system is now active at 28 hospital locations and supports medical teams around the clock in analysing X-ray and CT images.
SM004 Aidoc Hartford HealthCare and Aidoc Partnership The implementation moved from kickoff to go-live in just three weeks.
SM005 U.S. Food and Drug Administration AI-Enabled Medical Devices The FDA AI-enabled medical-device list shows radiology as the densest category on the current roster.
SM006 U.S. Food and Drug Administration Artificial Intelligence/Machine Learning (AI/ML)-Based Software as a Medical Device (SaMD) Action Plan The action plan specifically references AI/ML in radiological imaging and workflow automation.
SM007 U.S. Food and Drug Administration Predetermined Change Control Plans for Machine Learning-Enabled Device Software Functions (Draft Guidance) Predetermined change control plans are the FDA pathway for managing future machine-learning software modifications.
SM008 U.S. Food and Drug Administration 510(k) Premarket Notification database A 510(k) is a premarket submission made to FDA to demonstrate substantial equivalence to a legally marketed device.
SM009 U.S. Food and Drug Administration DEN170073 De Novo review for Viz LVO Viz.ai’s original LVO software entered through a De Novo rather than predicate-based 510(k) pathway.
SM010 U.S. Food and Drug Administration K231631 510(k) clearance letter for BriefCase-Quantification K231631 cleared Aidoc’s BriefCase-Quantification for coronary artery calcification quantification.
SM011 Centers for Medicare & Medicaid Services NHE Fact Sheet NHE grew 7.2% to $5.3 trillion in 2024 ... hospital expenditures grew to $1,634.7 billion and physician and clinical services to $1,109.7 billion.
SM012 American Hospital Association Fast Facts on U.S. Hospitals, 2026 There are 6,100 hospitals in the United States.
SM013 World Health Organization Cancer fact sheet Cancer is a leading cause of death worldwide, accounting for nearly 10 million deaths in 2022.
SM014 National Cancer Institute Cancer Statistics In 2025, an estimated 2,041,910 new cases of cancer will be diagnosed in the United States.
SM015 AAMC Physician Workforce Projections Annual physician workforce supply & demand projections for primary & specialty care in the US.
SM016 AAMC Summary Report: The Complexities of Physician Supply and Demand: Projections From 2021 to 2036 Physician demand is projected to continue to grow faster than supply, leading to a total projected shortage of between 13,500 and 86,000 physicians by 2036.
SM017 Harvey L. Neiman Health Policy Institute New Studies Shed Light on the Future Radiologist Workforce Shortage by Projecting Future Radiologist Supply and Demand for Imaging The present radiologist shortage is projected to persist unless steps are taken to grow the workforce and/or decrease per person imaging utilization.
SM018 American College of Radiology The Radiologist Shortage: A Workforce Update from HPI Changes in the practice landscape that have grown out of necessity with economic and regulatory pressures are creating a difficult environment for radiologists to thrive in.
SM019 PubMed Central Reimbursement in the age of generalist radiology artificial intelligence We argue that generalist radiology artificial intelligence challenges current healthcare reimbursement frameworks.
SM020 PubMed Central Services and payments associated with the medicare new technology add-on payment program The NTAP literature illustrates how difficult it is for novel technologies to convert regulatory novelty into durable reimbursement.
SM021 dotmed.com AI for pulmonary embolism detection shows high agreement with radiologists in real-world study Among confirmed pulmonary embolism cases, 15% were identified by radiologists but missed by the algorithm.
SM022 Radiology Business Radiology dominates FDA-cleared AI, but reimbursement lags far behind As of January 2026, there will only be two CPT category 1 payment codes for newer AI, despite there being hundreds of FDA-cleared medical imaging algorithms.
SM023 Aidoc Aidoc Foundation Model page CARE™ powered applications are delivered through aiOS™, the world’s most deployed clinical AI platform in 150+ health systems.
SM024 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses As hospitals seek broader, system-wide solutions, the market is shifting toward clinical AI deployed across entire health systems.
SM025 PR Newswire Aidoc Secures $150M for CARE, its Healthcare Foundation Model, to Transform Clinical Decision-Making for 100 Million Patients Aidoc currently supports care for more than 45 million patients annually across 150+ health systems, growing to 100 million in three years.
SM026 MarketsandMarkets Radiology AI Market Report 2025-2030, By Offering, Function, and Geo The global radiology AI market is projected to reach USD 2.27 billion by 2030, up from USD 0.76 billion in 2025, growing at a CAGR of 24.5%.
SM027 Emergen Research Artificial Intelligence (AI) in Radiology Market Size, Share, Trend Analysis by 2034 The Hospitals category captured the highest share in 2024 at about 55% of the total global market.
SP001 Aidoc AI Empowering Radiologists
SP002 Aidoc One integration. Every workflow. Proven at scale.
SP003 Aidoc Care Coordination Platform
SP004 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SP005 Viz.ai Viz.ai homepage
SP006 Viz.ai Viz.ai Raises $100 Million in Series D Funding
SP007 Enlitic Radiology
SP008 Harrison.ai Significant breakthrough in UK’s fight against lung cancer
SP009 Nano-X Imaging Ltd. Nanox Announces Fourth Quarter 2025 Financial Results and Provides Business Updates
SP010 Nanox Nanox homepage
SP011 PathAI Introducing AISight, the Digital Pathology Image Management System from PathAI
SP012 Roche Roche to acquire PathAI
SP013 Paige AI to solve cancer’s most critical issues
SP014 Tempus AI Tempus Announces the Acquisition of Paige
SP015 Microsoft Precision Imaging Network
SP016 Qure.ai Qure.ai joins Nuance Precision Imaging Network
SP017 Sectra Sectra One Cloud and Enterprise Imaging
SP018 AGFA HealthCare Enterprise Imaging Platform
SP019 FUJIFILM Healthcare Americas Synapse Artificial Intelligence Orchestrator
SP020 GE HealthCare GE HealthCare accelerates AI model development and deployment with launch of Edison integration to ACR AI-LAB
SP021 GE HealthCare GE HealthCare to acquire Intelerad, advancing cloud-enabled enterprise imaging across care settings
SP022 American College of Radiology AI Best Practices in Radiology — ARCH-AI
SP023 Radiology Business AI platforms evolving beyond marketplaces As currently, the ROI just doesn't make sense.
SP024 U.S. Food & Drug Administration K251406 BriefCase-Triage 510(k) clearance letter
SP025 PR Newswire The Era of Clinical AI Has Arrived: Trusted by Leading Health Systems, Aidoc's Platform Brings AI to the Heart of Patient Care
SP026 AWS Marketplace Aidoc clinical AI platform listing
SI001 Aidoc Aidoc $110 million funding announcement
SI002 PR Newswire The Era of Clinical AI Has Arrived: Trusted by Leading Health Systems, Aidoc's Platform Brings AI to the Heart of Patient Care
SI003 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SI004 Goldman Sachs Asset Management Aidoc Raises $150M Series E Led by Goldman Sachs
SI005 CTech by Calcalist New funding fuels rapid expansion of medical AI tools for hospitals worldwide
SI006 Globes AI medical decision co Aidoc raises $150m
SI007 Aidoc One integration. Every workflow. Proven at scale.
SI008 Aidoc AI Empowering Radiologists
SI009 Aidoc Care Coordination Platform
SI010 AWS Marketplace Aidoc clinical AI platform listing
SI011 ITQlick Aidoc - Always-on AI for Radiology Pricing Comparison
SI012 Aidoc Improving hospital length of stay with clinical AI
SI013 U.S. Food & Drug Administration K251406 BriefCase-Triage 510(k) clearance letter
SI014 Radiology Business AI platforms evolving beyond marketplaces As currently, the ROI just doesn't make sense.
SI015 American College of Radiology AI Best Practices in Radiology — ARCH-AI
SI016 GE HealthCare GE HealthCare to acquire Intelerad, advancing cloud-enabled enterprise imaging across care settings
SI017 Nano-X Imaging Ltd. Nanox Announces Fourth Quarter 2025 Financial Results and Provides Business Updates
SI018 Tempus AI Tempus Announces the Acquisition of Paige
SI019 Roche Roche to acquire PathAI
SI020 Viz.ai Viz.ai Raises $100 Million in Series D Funding
SI021 Microsoft Precision Imaging Network
SI022 Qure.ai Qure.ai joins Nuance Precision Imaging Network
SI023 Aidoc Aidoc homepage
SI024 PR Newswire Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SI025 FUJIFILM Healthcare Americas Synapse Artificial Intelligence Orchestrator
SI026 Sectra Sectra One Cloud and Enterprise Imaging
SI027 Aidoc Read Our Latest News and Updates | Aidoc
SI028 Aidoc Sol Radiology and Aidoc Partner to Advance AI-Powered Imaging Across Southern California - Aidoc | Clinical AI
SI029 Aidoc Isala Hospital Leads the Way with First AI-Powered Pulmonary Embolism Response in the Netherlands - Aidoc | Clinical AI
SI030 Aidoc Aidoc Introduces New Comprehensive Triage Solution for Europe Built on Foundation Model AI - Aidoc | Clinical AI
SI031 PR Newswire Former AMA President Dr. Jesse Ehrenfeld Joins Aidoc as Chief Medical Officer Amid Growing Health System Adoption of Clinical AI
SE001 Aidoc aiOS™ | End-To-End Clinical AI Platform
SE002 Aidoc Aidoc Foundation Model / CARE overview
SE003 Aidoc Aidoc Secures FDA Clearance for Healthcare’s First Comprehensive Foundation Model AI
SE004 U.S. Food & Drug Administration K252970 BriefCase-Triage: CARE Multi-triage CT Body
SE005 Aidoc Neuro solutions / Full Brain AI
SE006 Aidoc VTE clinical AI solutions
SE007 Aidoc Aortic solutions
SE008 Aidoc Aidoc Security & Privacy
SE009 Aidoc Quality & Compliance
SE010 Aidoc Care Coordination Solutions
SE011 Aidoc Aidoc and AWS Partnership
SE012 Aidoc BRIDGE framework
SE013 Aidoc Aidoc Job Opportunities
SE014 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SE015 PR Newswire Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SE016 MedTech Dive Aidoc raises $150M to advance clinical AI foundation model
SE017 HIT Consultant Aidoc Secures $150M to Accelerate Enterprise-Scale Clinical AI
SE018 Fierce Healthcare Clinical AI company Aidoc lands $150M backed by General Catalyst, Nvidia's venture arm
SE019 CTech / Calcalist FDA clears Aidoc’s foundation-model AI for broad clinical triage
SE020 Applied Radiology Aidoc Earns FDA Breakthrough Device Designation for Multi-Condition Clinical AI Platform
SE021 Diagnostic Imaging FDA Clears CT-Based AI Triage Platform from Aidoc
SE022 510k Database Aidoc Medical, Ltd. - 34 FDA 510(k) Radiology Devices (Latest 2026)
SE023 Healthcare IT Today Aidoc Secures $150M for CARE, its Healthcare Foundation Model, to Transform Clinical Decision-Making for 100 Million Patients
SE024 Aidoc Hartford HealthCare and Aidoc Partner to Transform Patient Care with Enterprise AI
SE025 Aidoc 50 Facilities, 4 Months: Mercy’s Bold AI Deployment Story
SE026 Aidoc Asklepios Successfully Completes Comprehensive AI Rollout in Radiology: 28 Hospitals Using Aidoc to Support Patient Care
SE027 HIT Consultant Hartford HealthCare Deploys Aidoc’s AI-Enabled Solutions Across Enterprise
SU001 Aidoc Hear Healthcare AI Success Stories from Hospitals Worldwide
SU002 Aidoc Hartford HealthCare and Aidoc Partner to Transform Patient Care with Enterprise AI
SU003 HIT Consultant Hartford HealthCare Deploys Aidoc’s AI-Enabled Solutions Across Enterprise
SU004 Aidoc Change at the Speed of Trust: 4 Takeaways from Aidoc's Clinical AI Partnership with Hartford Healthcare
SU005 Aidoc Real World Emergency AI Triage in Action at Mercy
SU006 Aidoc 50 Facilities, 4 Months: Mercy’s Bold AI Deployment Story
SU007 Aidoc Asklepios Successfully Completes Comprehensive AI Rollout in Radiology: 28 Hospitals Using Aidoc to Support Patient Care
SU008 Sutter Health Sutter Health and Aidoc Team Up to Transform Patient Care with Advanced Clinical AI
SU009 CaseStudies.com Case Study: Yale New-Haven Hospital improves PE response and advanced therapy use with Aidoc
SU010 Aidoc AI-Powered Hub-and-Spoke Stroke Care at Renown Health & Carson Tahoe Health
SU011 Aidoc Temple Health's CEO on Why AI Is a Strategic Imperative
SU012 Healthcare IT Today Aidoc Secures $150M for CARE, its Healthcare Foundation Model, to Transform Clinical Decision-Making for 100 Million Patients
SU013 Fierce Healthcare Clinical AI company Aidoc lands $150M backed by General Catalyst, Nvidia's venture arm
SU014 MedCity News Aidoc Rakes In $150M for Its Clinical Decision Support AI
SU015 HLTH Aidoc Secures $150M to Expand Clinical AI Platform Beyond Radiology
SU016 FeaturedCustomers 51 Aidoc Customer Reviews & References
SU017 ITQlick Aidoc - Always-on AI for Radiology Reviews 2026: Real Pros, Cons & Expert Value Verdict High initial investment cost, presenting a barrier for smaller clinics to adopt.
SU018 Aidoc aiOS™ | End-To-End Clinical AI Platform
SU019 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SU020 PR Newswire Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SU021 Aidoc VTE Clinical AI Solutions
SU022 Aidoc AI Aortic Solutions
SU023 Aidoc Neuro solutions / Full Brain AI
SU024 Aidoc Aidoc & Mercy: AI at Million-Image Scale
SU025 Aidoc Aidoc & Mercy: Clinical AI Implemented at Unprecedented Speed
SR001 Aidoc Quality & Compliance Certified to MDSAP, compliant with FDA’s Quality System Regulation (21 CFR Part 820), certified to ISO 13485:2016 and EU 2017/745 (MDR).
SR002 Aidoc Aidoc Security & Privacy
SR003 Aidoc Data Privacy Framework Notice
SR004 Aidoc Clinical AI Partnerships & Integrations | Aidoc’s Unified Platform
SR005 Aidoc A Roadmap for Scalable, Responsible AI Adoption in Healthcare
SR006 Aidoc Aidoc and AWS Partnership
SR007 Aidoc How Foundation Models Are Transforming Clinical AI
SR008 Aidoc Aidoc Secures New FDA Clearance
SR009 Aidoc Dr. Jesse Ehrenfeld Joins Aidoc as Chief Medical Officer
SR010 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SR011 Aidoc Isala Hospital Leads the Way with First AI-Powered Pulmonary Embolism Response in the Netherlands
SR012 Aidoc Asklepios Successfully Completes Comprehensive AI Rollout in Radiology: 28 Hospitals Using Aidoc to Support Patient Care
SR013 Aidoc Sol Radiology and Aidoc Partner to Advance AI-Powered Imaging Across Southern California
SR014 Aidoc Temple Health Delivers More Timely Care with AI-Driven Triage
SR015 Aidoc Aidoc & Mercy: AI at Million-Image Scale
SR016 Aidoc How Aidoc AI Transformed Stroke Care Across 30 Hospitals Across Ochsner Health
SR017 Aidoc From Hours to Minutes: How Aidoc Transformed PE Care at Lehigh Valley Health Network
SR018 Advocate Health Advocate Health Deploys AI Solution to Redefine Diagnostic Excellence through Agreement with Aidoc
SR019 Novant Health Aidoc partners with Novant Health, providing imaging AI to expedite treatment for patients in the emergency department
SR020 National Library of Medicine Decreased Hospital Length of Stay for ICH and PE after Adoption of an Artificial Intelligence-Augmented Radiological Worklist Triage System
SR021 The Royal College of Radiologists Aidoc ICH
SR022 PR Newswire Aidoc Announces Collaboration with AWS to Advance Clinical AI Foundation Models, Transforming Healthcare at Scale
SR023 Radiology Business Aidoc scores ‘significant’ investment from Amazon, seeks to flesh out radiology foundation model
SR024 Forbes AWS and Aidoc’s Collaboration Is Making Waves In Clinical AI
SR025 U.S. Food and Drug Administration AI-Enabled Medical Devices
SR026 U.S. Food and Drug Administration Artificial Intelligence-Enabled Device Software Functions: Lifecycle Management and Marketing Submission Recommendations
SR027 U.S. Food and Drug Administration Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence-Enabled Device Software Functions
SR028 U.S. Food and Drug Administration Good Machine Learning Practice for Medical Device Development
SR029 Chambers and Partners Healthcare AI 2025 | Global Practice Guides
SR030 U.S. Department of Health and Human Services Resolution Agreements
SR031 U.S. Department of Health and Human Services HHS’ Office for Civil Rights Settles Four HIPAA Security Rule Ransomware Investigations
SR032 U.S. Department of Health and Human Services Enforcement Data
SR033 American College of Radiology Reimbursement for AI in Radiology: Practices and Considerations
SR034 Radiology Business Radiology dominates FDA-cleared AI, but reimbursement lags far behind
SR035 HIPAA Journal HIPAA Violation Cases - Updated 2026
SR036 Epic Systems Artificial Intelligence | Epic
SR037 Microsoft AI Solutions Optimized for Epic | Microsoft for Healthcare
SR038 Oracle Health Oracle Health Clinical AI Agent
SR039 Diagnostic and Interventional Radiology Bias in artificial intelligence for medical imaging: fundamentals, detection, avoidance, mitigation, challenges, ethics, and prospects
SR040 National Library of Medicine Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification
SV001 PR Newswire Aidoc raises $110M to Address the Increasing Challenges Facing Health Systems by Using Artificial Intelligence
SV002 MobiHealthNews Aidoc raises $110M to expand AI-enabled imaging platform
SV003 Aidoc Aidoc Raises $150 Million Series E Led by Goldman Sachs to Scale Clinical AI for Earlier, Safer Diagnoses
SV004 Aidoc Dr. Jesse Ehrenfeld Joins Aidoc as Chief Medical Officer
SV005 Aidoc Aidoc Secures New FDA Clearance
SV006 Aidoc Asklepios Successfully Completes Comprehensive AI Rollout in Radiology: 28 Hospitals Using Aidoc to Support Patient Care
SV007 Advocate Health Advocate Health Deploys AI Solution to Redefine Diagnostic Excellence through Agreement with Aidoc
SV008 PR Newswire Aidoc Announces Collaboration with AWS to Advance Clinical AI Foundation Models, Transforming Healthcare at Scale
SV009 CTech Aidoc raises $150M with Nvidia backing as AI pushes faster diagnosis
SV010 Fierce Healthcare Clinical AI company Aidoc lands $150M backed by General Catalyst, Nvidia's venture arm
SV011 Business Wire Viz.ai Raises $100 Million in Series D Funding, Led by Tiger Global and Insight Partners at $1.2 Billion Valuation
SV012 Abridge Abridge Series E Announcement and More
SV013 Roche Roche enters into a definitive merger agreement to acquire PathAI to transform AI-driven diagnostics
SV014 PR Newswire PathAI Announces Completion of $165 Million Financing for Advancing Medicine with AI-powered Pathology
SV015 Tempus Tempus Announces the Acquisition of Paige
SV016 CompaniesMarketCap Tempus AI (TEM) - Market capitalization
SV017 CompaniesMarketCap RadNet (RDNT) - Market capitalization
SV018 CompaniesMarketCap Phreesia (PHR) - Market capitalization
SV019 CompaniesMarketCap Health Catalyst (HCAT) - Market capitalization
SV020 CompaniesMarketCap GE HealthCare Technologies (GEHC) - Market capitalization
SV021 U.S. Securities and Exchange Commission Tempus AI companyfacts JSON
SV022 U.S. Securities and Exchange Commission RadNet companyfacts JSON
SV023 U.S. Securities and Exchange Commission Phreesia companyfacts JSON
SV024 U.S. Securities and Exchange Commission Health Catalyst companyfacts JSON
SV025 U.S. Securities and Exchange Commission GE HealthCare companyfacts JSON
SV026 Rock Health 2025 year-end digital health funding overview: A tale of two markets
SV027 Rock Health Q1 2026 funding overview: Capital continues concentrating and four other market signals
SV028 Cooley Q4 2025 Venture Financing Report: Up and Flat Rounds Increased; Recapitalization, Pay to Play and Redemption Decreased
SV029 Radiology Business Radiology dominates FDA-cleared AI, but reimbursement lags far behind
SV030 American College of Radiology Reimbursement for AI in Radiology: Practices and Considerations
SV031 Aidoc Aidoc and AWS Partnership