CASSIE

Cassie is a cloud-based software-as-a-medical-device (SaMD) that analyzes photoplethysmogram (PPG) data collected during sleep to automatically classify sleep stages and detect respiratory events. It is intended for use by healthcare professionals to aid in the evaluation and management of sleep disorders in adults, producing metrics such as total sleep time, sleep efficiency, and apnea-hypopnea index (AHI) when oximetry data is available.

Both Cassie and the predicate device (SleepImage) are cloud-based SaMD applications that accept PPG and optional SpO2 inputs and produce sleep architecture and respiratory event outputs using cardiopulmonary physiological analysis. The key difference is that Cassie uses machine learning algorithms whereas the predicate uses automatic analysis, but both process the same physiological data to achieve equivalent clinical endpoints. Cassie's inputs and outputs are a subset of the predicate's broader capabilities.

Software development followed IEC 62304:2006 (Medical Device Software – Software Life-Cycle Processes). Verification and validation testing were conducted per FDA guidance on cybersecurity in medical devices (June 2025), device software content (June 2023), and SaMD clinical evaluation (IMDRF/SaMD WG/N41FINAL:2017). Clinical validation used polysomnography (PSG) as the reference standard, scored per AASM guidelines by certified technologists.

Cassie is substantially equivalent to the predicate SleepImage (K182618) because both devices share the same intended use (analyzing PPG data to aid sleep disorder diagnosis and management), employ similar technological principles (cloud-based analysis of cardiopulmonary markers), and demonstrate comparable clinical performance when validated against PSG. Key performance endpoints (AHI correlation 0.96, TST correlation 0.86, sensitivity/specificity for OSA detection) match or exceed predicate performance, confirming that the technological difference (machine learning vs. automatic analysis) does not raise new questions of safety or effectiveness.