Hypertension Prediction Index (HePI) Algorithm

The Hypertension Prediction Index (HePI) is a software algorithm that predicts the likelihood of future hypertensive events in adult patients (18+) receiving advanced hemodynamic monitoring during surgical or non-surgical procedures. It provides clinicians with a value from 0-100 updated every 20 seconds and triggers alerts when values exceed 85, intended for reference only to support clinical decision-making in critical care settings.

HePI maintains the same overall architecture, input processing, and output steps as the predicate HemoSphere Advanced Monitoring Platform with Acumen Hypotension Prediction Index (HPI). The key difference is that HePI predicts hypertensive events (MAP >115 mmHg or MAP increase >20% when >95 mmHg) rather than hypotensive events, and uses a slightly different machine learning model to generate the prediction index.

ANSI/AAMI/IEC 62304:2006/A1:2016 (Medical Device Software – Software Life Cycle Processes) and FDA's Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices (May 11, 2005).

HePI is substantially equivalent because it shares the same technological platform, architecture, and processing methodology as the predicate HPI device, with demonstrated sensitivity and specificity above 80% (actually 91.5-99.9% and 90.6-94.3% across datasets). Although the algorithm predicts a different physiological event (hypertension vs. hypotension) using a modified machine learning model, performance testing shows no new safety or effectiveness concerns, and both devices serve as adjunctive predictive indicators in the same clinical setting and patient population.