Introduction
Coronary artery calcium is one of the most reliable indicators of cardiovascular risk, and it is frequently visible on chest CT scans performed for entirely non-cardiac reasons. In routine practice, however, that incidental calcium is often not quantified or communicated, leaving preventable risk unaddressed.
The AI INFORM trial — presented at the 2026 Society of Cardiovascular Computed Tomography (SCCT) Annual Meeting by investigators from Brigham and Women’s Hospital and Mass General Brigham — examined whether notifying clinicians about AI-detected coronary calcium changes prescribing behavior. Using Nanox.AI’s FDA-cleared cardiac AI solution (HealthCCSng) to automatically flag calcium on existing scans, the study measured how notification affected the initiation or intensification of lipid-lowering therapy (LLT). This page summarizes the design, headline findings, and clinical relevance, and links to the full poster.
Key Takeaways
Study Overview
The AI INFORM trial (« Artificial Intelligence Detection of Incidental Coronary Artery Calcium to Enhance Cardiovascular Disease Prevention ») is a multi-site randomized trial designed to determine whether giving clinicians information about AI-detected coronary plaque increases the initiation or intensification of preventive lipid-lowering therapy. The initial single-center results from Mass General Brigham were presented at SCCT 2026.
- Design: Randomized comparison of a notification arm versus usual care.
- Population (single-center analysis): 660 patients — 325 in the notification arm (49.2%) and 335 in the usual-care arm (50.8%).
- AI CAC determination: Cloud-based, AI-powered detection using Nanox.AI HealthCCSng (FDA-cleared).
- Primary focus: Initiation or intensification of lipid-lowering therapy (LLT) at 6 and 12 months.
Key Findings
Across both time points, clinician notification of AI-detected coronary calcium was associated with substantially higher use of preventive lipid-lowering therapy.
6-Month Outcomes (notification vs usual care):
- LLT initiation or intensification: 138 (42.5%) vs 34 (10.1%), p<0.001
- Statin initiation or intensification: 134 (41.2%) vs 32 (9.6%), p<0.001
12-Month Outcomes, cumulative (notification vs usual care):
- LLT initiation or intensification: 204 (62.7%) vs 64 (19.1%), p<0.001
- Statin initiation or intensification: 191 (58.8%) vs 56 (16.7%), p<0.001
Rates of downstream non-invasive imaging and invasive coronary angiography were low and similar between arms, suggesting the notification prompted preventive medical therapy rather than a cascade of additional testing. The reliability of the underlying detection is examined in a companion multi-site AI vs. expert agreement analysis.
Note: These are initial single-center results presented in abstract/poster form. Full peer-reviewed publication of the complete multi-site dataset may refine these figures.
Why This Matters for Cardiovascular Prevention
Incidental coronary calcium represents one of the largest untapped opportunities in preventive cardiology. When that signal is surfaced and communicated, clinicians can act on it.
- No additional imaging or radiation — the AI analyzes scans that have already been performed.
- Fits existing workflows — findings are surfaced through AI analysis of routine CT scans rather than a separate imaging protocol.
- Scalable across large populations — opportunistic screening can reach patients who would never be referred for dedicated cardiac CT.
- Aligned with preventive-care goals — earlier initiation of statins and other lipid-lowering therapy is central to reducing cardiovascular events.
The AI INFORM trial reinforces a broader shift: using artificial intelligence to move from a missed finding to a clinical action.
FAQs
AI INFORM is a randomized trial evaluating whether notifying clinicians about AI-detected incidental coronary artery calcium leads to greater initiation or intensification of preventive lipid-lowering therapy. It was presented at SCCT 2026 by investigators from Brigham and Women’s Hospital and Mass General Brigham.
Coronary calcium was identified automatically on routine, non-cardiac CT scans using Nanox.AI’s FDA-cleared cardiac solution, HealthCCSng — a cloud-based, AI-powered platform.
Clinician notification was associated with significantly higher use of lipid-lowering therapy: 42.5% versus 10.1% at 6 months and 62.7% versus 19.1% at 12 months compared with usual care.
Coronary artery calcium is a strong marker of cardiovascular risk. Because it is often visible on scans performed for other reasons, detecting and communicating it creates an opportunity for earlier preventive treatment.
Portail client