The Food and Drug Administration published a final order in the Federal Register on Sept. 11, 2026, classifying cardiovascular machine learning-based notification software as Class II with special controls and adding 21 CFR 870.2380.
FDA defines the type as software that uses machine learning on one or more non-invasive physiological inputs from routine care to suggest the likelihood of a single cardiovascular condition for referral or further testing. It is not intended to deliver diagnostic-quality output and is not intended to identify or detect arrhythmias.
The order is effective Sept. 11, 2026. FDA says the classification was already applicable on Aug. 3, 2023, when it granted Viz.ai Inc.'s De Novo request for Viz HCM (DEN230003; product code QXO).
Per the De Novo classification order, Viz HCM analyzes compatible 12-lead electrocardiogram recordings for signs associated with hypertrophic cardiomyopathy in patients 18 and older, not patients with implanted pacemakers, and must not replace full clinical evaluation or be used to rule out HCM follow-up.
Special controls listed in the final order include clinical performance testing on real-world data with an independent, geographically diverse three-site test set; software verification, validation, and hazard analysis; human-factors assessment for misinterpretation risk; and labeling that warns against relying on a negative finding to skip follow-up.
FDA has not exempted the type from 510(k). Once Class II via De Novo, the device type can serve as a predicate so later substantially equivalent products may use the 510(k) path instead of another De Novo or premarket approval.
Independent: Regulatory Focus (RAPS) noted the Sept. 11 classification among that week's FDA device orders. Innolitics separately describes 21 CFR 870.2380 as the Viz HCM De Novo regulation later used by follow-on cardiovascular machine-learning notification clearances. Treat product-specific performance claims beyond the order text as unverified here.