研究目的
To propose a novel approach for accurate monitoring of heart rate (HR) using Photoplethysmography (PPG) signals in the presence of Motion Artifacts (MA) during radical exercises.
研究成果
The proposed framework effectively estimates HR from wrist-type PPG signals during physical exercises, achieving an average absolute error of 1.66 BPM. It demonstrates robustness in removing MA and offers a promising approach for continuous HR monitoring in WBANs.
研究不足
The study focuses on wrist-type PPG signals during physical exercises, which may not cover all scenarios of MA interference. The algorithm's performance in clinical scenarios with small MAs was not extensively tested.
1：Experimental Design and Method Selection:
The framework consists of signal decomposition for denoising using PCA, spare signal reconstruction (SSR), peak detection and tracking, and SVM classifier for accurate estimation of HR.
2：Sample Selection and Data Sources:
Datasets obtained from 2015 IEEE Signal Processing CUP, involving 12 subjects running on a treadmill with varying speeds.
3：List of Experimental Equipment and Materials:
Wrist-type PPG signals and acceleration signal from a 3-axis accelerometer.
4：Experimental Procedures and Operational Workflow:
Signal decomposition, SSR, peak detection and tracking, and SVM classification.
5：Data Analysis Methods:
Performance assessed using average absolute error between original HR and estimated HR.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
