One Scan, 14 Diagnoses: AI Jumps the Radiology Queue

Aidoc's CARE model is the first FDA-cleared AI to detect 14 acute conditions from a single abdominal CT scan

6 Mar 2026

The US Food and Drug Administration has cleared an artificial intelligence tool that can screen a single abdominal CT scan for 14 acute conditions simultaneously, a regulatory milestone that signals a shift in how emergency imaging is managed in American hospitals.

Aidoc, an Israeli-founded medical AI company, received clearance on 21 January for its Clinical AI Reasoning Engine, known as CARE. The system automatically reviews incoming scans and ranks the most urgent cases at the top of a radiologist's queue, before any human review takes place. Conditions it can flag include appendicitis, bowel obstruction, aortic dissection, pelvic fracture, and kidney stones.

The clearance covers 14 indications in total. In the FDA-reviewed clinical study, the 11 newly approved conditions achieved a mean sensitivity of 97 per cent and mean specificity of 98 per cent. The system also produces fewer false alerts than older, single-condition tools, a factor that clinical teams have historically cited as a barrier to adopting AI in radiology workflows.

The architecture behind CARE distinguishes it from prior approvals. Most cleared radiology AI products are designed to detect a single condition. CARE's clearance as a multi-indication foundation model represents the first time the FDA has sanctioned this type of broad-scope clinical AI for double-digit acute conditions in one pass.

"This FDA clearance, combining unprecedented breadth and accuracy, marks a new era for clinical AI," said Elad Walach, Aidoc's chief executive.

Heidi Beilis, chief medical officer of diagnostics at WellSpan Health, described the tool as a fundamental shift in how radiology departments manage workflow and accelerate time-to-diagnosis for acute conditions.

Aidoc currently operates across more than 1,600 medical centres and has processed over 100 million patient cases, providing an established deployment base for the new model. The company has stated it plans to extend CARE across all CT and X-ray workflows within 18 months, with automated radiology report drafting as the next development target.

US emergency departments process thousands of CT scans daily under conventional first-come, first-served protocols. The clearance arrives as health systems face growing pressure from rising imaging volumes and staffing constraints, with AI-assisted triage moving from pilot programmes into standard clinical infrastructure.