Application of Improved Wavelet Transform and BiLSTM in PPG Signal Processing
This paper discusses the critical role of Electrocardiogram(ECG)and Photoplethysmog-raphy(PPG)signals in modern medical and health monitoring,particularly for real-time cardiac health and blood flow dynamics.ECG and PPG signals are susceptible to noise interference during daily activi-ties and exercise,highlighting the need for effective noise removal techniques.The paper proposes an improved wavelet transform denoising algorithm combined with a BiLSTM(Bidirectional Long Short-Term Memory)model for signal processing.Experimental results show that the improved wavelet transform method outperforms traditional methods in noise reduction and signal detail preservation.The BiLSTM model,leveraging both forward and backward information,enhances heart rate estimation ac-curacy.Based on ISPC database experimental data,the BiLSTM-PPG model achieves the lowest mean absolute error(MAE)of 2.09 BPM,surpassing TROIKA and Deep PPG models.
Photoplethysmography(PPG)wavelet transformBidirectional Long Short-Term Memory(BiLSTM)heart rate measurement
万方数据
维普
