Dynamic HRV Monitoring and Machine Learning Predict NYHA Improvements in Acute Heart Failure Patients

Ying Shi, Xiu Zhang, Chenbin Ma, Yue Zhang, Zhicheng Yang, Wei Yan, Muyang Yan, Qing Zhang, Zhengbo Zhang

Abstract: Highlights•ΔHRV (discharge-admission) predicts acute-to-stable transition in heart failure.•Wearable ECG + machine learning achieves 70+% accuracy in NYHA improvement prediction.•SDNN and SD2 emerge as key biomarkers for autonomic recovery in HF.•Real-time ΔHRV monitoring outperforms NT-proBNP in accessibility and sensitivity.•AI-driven monitoring prioritizes high-risk HF patients.