FDA classification of ML-based cardiovascular notification software

Current understanding

FDA final rule classifies cardiovascular machine-learning-based notification software as Class II (moderate risk) subject to special controls under the cardiovascular device regulations. For developers, this establishes a defined 510(k) pathway with device-specific performance, labeling, and cybersecurity controls rather than requiring de novo or PMA review. For clinicians and users, notifications generated by these tools carry a federally recognized risk classification and disclosure baseline; status is a final rule.

Evidence log

• 2026-09-11 — Medical Devices; Cardiovascular Devices; Classification of the Cardiovascular Machine Learning-Based Notification Software: cross-connection with fda-ml-quantitative-imaging-classification: Two parallel FDA final rules establishing Class II special-controls pathways for distinct ML medical device types (cardiovascular notification software; radiological quantitative imaging with PCCP), together forming an emerging device-by-device FDA approach to AI/ML software classification. (novelty: 2)

Open questions

Related