Aidoc vs Rad AI (2026)
The two most commonly shortlisted radiology AI purchases, and they do not overlap at all. Aidoc works on the image, running detection models across the imaging estate through an operating system that handles normalization, monitoring and governance, with a 2026 clearance covering many acute indications from one named foundation model. Rad AI never touches the image, generating the report impression from the radiologist's own dictated findings and learning their individual language from prior reports, with oversight built into the workflow itself. Detection changes which study gets read next; reporting changes how long each one takes to finish. If your backlog is turnaround time on urgent findings, Aidoc. If it is that radiologists are reading fast and reporting slowly, Rad AI is aimed at the actual bottleneck.
- It works on the image, running detection models across an imaging estate through an operating system that handles normalization, monitoring and governance for every model it runs.
- The January 2026 clearance covers many acute indications from a single named foundation model, with performance from a regulator reviewed pivotal study rather than a vendor benchmark.
- For a department whose problem is that urgent findings wait in the queue, detection and worklist reprioritisation is the intervention.
- It works on the report, generating the impression from the radiologist's own dictated findings and learning that radiologist's language from their prior reports.
- The workflow enforces oversight, since the model drafts and the radiologist signs, so no unread text reaches a released report.
- It publishes specific falsifiable operational figures with named baselines, including follow up completion, rather than adoption counts.
This comparison is published by AI Health Index, an independent research platform that compares healthcare AI vendors objectively. Aidoc and Rad AI are each graded against the same capability taxonomy, from each vendor's own public materials and the regulatory record, under the AI Health Index verification standard. No vendor pays for placement, and no vendor has reviewed this page. How this evidence is graded
Plain facts
Side by Side
Select any grade to read the note behind it.
Product updates
The latest logged changes from each vendor, newest first, each linked to its source.
Aidoc
-
Regulatory / FDA
Aidoc received FDA Breakthrough Device Designation for CARE Chest X-Ray Triage, built on its CARE foundation model, covering triage of pneumomediastinum and lobar or lung collapse.
Source -
Clinical evidence
Mercy and Aidoc released a white paper, commissioned by AVIA, reporting on five months of AI deployment across the Mercy health system.
Source -
Regulatory / FDA
The FDA cleared Aidoc's comprehensive abdomen CT triage solution, powered by its self developed CARE foundation model, bringing 11 newly cleared indications together with three previously cleared into a single 14 indication workflow.
Source
Each vendor in one paragraph
Written to be quoted whole. Each paragraph opens the AI Health Index profile of the vendor: what the product is, who buys it and what it does.
Aidoc sells clinical AI for imaging triage and follow up to health systems, delivered through aiOS, an enterprise platform that runs Aidoc's own FDA cleared algorithms alongside vetted partner algorithms and carries care coordination and patient management. Aidoc Medical was established in early 2016 in Tel Aviv and is led by cofounder and chief executive Elad Walach.
Source: AI Health Index, August 2026
Generative AI for radiology reporting and follow up. Rad AI Reporting is a cloud native reporting system with real time quality checks before sign off; Rad AI Impressions (Omni Impressions) drafts the report impression from the radiologist's dictated findings, customized to each radiologist's language from the site's de identified historical reports; Omni Unchanged carries stable findings forward from prior reports; Rad AI Continuity tracks follow up recommendations through the EHR and RIS.
Source: AI Health Index, August 2026
Questions buyers ask
How are Aidoc and Rad AI priced and sold?
Aidoc sells aiOS and its cleared modules to health systems under enterprise agreements through a demo, and maintains an AWS Marketplace listing, which makes the procurement channel public; the listing states that pricing depends on the duration and terms of the contract agreed with the vendor. Whether pricing runs per module, per study volume or per site, and how partner algorithms on aiOS are billed, is not stated.
Rad AI sells through agreements with radiology practices and health systems, and inquiries go through a demo request. From the AI Health Index, drawn from each vendor's own published materials and the regulatory record, verified August 4, 2026. No vendor pays for placement.
How do Aidoc and Rad AI handle patient data?
In Aidoc's case, the company's security and privacy documentation describes a new DICOM object created before upload carrying only non PHI data and a newly generated identifier; analysis runs on that object, and the identifier re-attaches PHI locally when results return.
For Rad AI, the security page describes a de identification pipeline specialized to radiology reports, and the FAQ says the models are trained on de identified historical reports, with no images, and customized for each site from three to five years of its de identified reports. From the AI Health Index, drawn from each vendor's own published materials and the regulatory record, verified August 4, 2026. No vendor pays for placement.
Which settings and specialties do Aidoc and Rad AI cover?
With Aidoc, the company reports use in more than 1,600 medical centers and names health systems including WellSpan, Mercy, Yale New Haven, St. Luke's and Renown. Validation sits at the level of each cleared indication, with its own FDA review, and the research library is organized by condition.
In Rad AI's case, the products cover report generation, impression drafting and follow up of incidental findings, sold to hospital radiology departments, physician owned practices and teleradiology groups; named users include AdventHealth, Advocate Health, Emory, Inland Imaging, LucidHealth and Trinity Health, and four health systems are strategic investors. From the AI Health Index, drawn from each vendor's own published materials and the regulatory record, verified August 4, 2026. No vendor pays for placement.
How much do Aidoc and Rad AI leave to the clinician?
Aidoc flags suspected acute findings, reorders the worklist and pushes a notification with a preview image to a desktop widget; results can also go to the native worklist, the reporting system or PACS as a separate series. The K252970 indication states the limits: analysis runs in parallel with the standard read, preview images are for notification only, the device is not for diagnosis, and notified clinicians must view the full images.
The company's FAQ states that a licensed radiologist must always review the AI generated impression, edit it as needed and sign the final report; Rad AI Reporting runs real time quality checks before sign off, and the security page describes human in the loop quality assurance since 2019. The company also states the tool catches clinically significant errors in the radiologist's dictation in about 5 percent of reports. From the AI Health Index, drawn from each vendor's own published materials and the regulatory record, verified August 4, 2026. No vendor pays for placement.
Related comparisons
Other published head to head assessments involving these vendors or their closest peers.
These are complementary and a large department will plausibly run both, so the useful question is sequencing rather than selection: detection changes which study is read next, reporting changes how long each study takes to finish. Rad AI holds no clearance for any product in its portfolio, which is consistent with reporting assistance rather than diagnosis but should be checked against how your department intends to use carried forward prior findings. Neither publishes pricing. Neither publishes subgroup performance, and for detection models trained on particular scanner and protocol mixes, local validation before go live is the diligence that gets skipped most often.