Congestive Heart Failure Staging on a Smartphone

Congestive heart failure (CHF) is a chronic condition where the heart weakens and struggles to pump enough blood. It progresses through four stages based on New York Heart Association criteria, with early stages often asymptomatic. Without intervention, patients may unknowingly advance to more severe stages. Regular CHF monitoring is crucial but typically requires hospital visits, making frequent assessments inconvenient. This letter proposes an AI-based CHF staging method that runs on smartphones, enabling patients to monitor their condition at home. The approach leverages photoplethysmography and skin impedance at acupoints to train a deep learning model. Using data from 479 CHF patients, the model achieves 92% accuracy in predicting CHF stages. This smartphone-based AI solution offers a convenient, noninvasive alternative for CHF monitoring, reducing hospital dependence while improving disease management.