AI-driven non‑invasive cardiac monitoring

In late July 2026, researchers at Ateneo de Manila University unveiled an artificial‑intelligence model that estimates the cardiac index—a key measure of heart‑pump performance—using data from simple skin‑adhesive sensor stickers. The system achieved 97.78 % classification accuracy, suggesting it can deliver reliable cardiac assessments without specialized hemodynamic equipment, potentially expanding cardiovascular monitoring in resource‑limited settings, especially in the Philippines.

A few days later, a multidisciplinary team published a broader AI foundation model in *Nature Communications* that analyzes routine sleep‑study recordings to predict long‑term health risks, including heart disease, cognitive decline and five‑year mortality. The study highlighted five risk categories, with the highest‑risk group facing a doubled five‑year mortality risk. The report also reiterated the Ateneo‑based cardiac‑index model, underscoring its continued development and its role as a low‑cost alternative for detailed heart‑function monitoring in underserved clinics.

Together, the snapshots trace a rapid progression from a focused, non‑invasive cardiac‑index tool toward a wider AI‑enabled health‑risk platform that leverages everyday sensor data, reflecting growing interest in affordable, AI‑powered cardiovascular diagnostics.