Abstract
Wearable health monitoring enhances real-time physiological tracking and anomaly detection. However, Photoplethysmography (PPG) and Heart Rate Variability (HRV) signals often suffer from noise and motion artifacts. This study proposes a machine learning framework integrating noise reduction techniques for improved signal classification. Random Forest (RF), Support Vector Machine (SVM), and Long Short-Term Memory (LSTM) models are evaluated, with LSTM achieving the highest accuracy (93.5%) post-noise reduction. Comparative analysis highlights deep learning's superiority in biomedical signal processing. These findings support the development of intelligent wearable health solutions for early disease detection and personalized healthcare.
| Original language | English |
|---|---|
| Pages (from-to) | 106-112 |
| Number of pages | 7 |
| Journal | Procedia Computer Science |
| Volume | 284 |
| DOIs | |
| Publication status | Published - 2026 |
| Event | 8th Information Systems International Conference, ISICO 2025 - Hybrid, Bali, Indonesia Duration: 4 Aug 2025 → 6 Aug 2025 |
Keywords
- Anomaly detection
- HRV
- Machine learning
- Noise reduction
- PPG
- Wearable sensors
Fingerprint
Dive into the research topics of 'Optimizing health prediction using wearable sensor data: Machine learning for anomaly detection in PPG and HRV signals'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver