Removal of Motion Artifacts From the PPG Signal Using Attentive Generative Adversarial Networks With Dual Discriminator
- 14 January 2025
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-10
Abstract
With the widespread integration of smartwatches and fitness trackers into daily life, photoplethysmography (PPG) signals have emerged as one of the most popular bio-signals. However, motion artifacts often affect these signals, diminishing their practical applicability. This study proposes using generative adversarial networks (GANs) to remove these motion artifacts from the PPG signals without additional motion data from accelerometers or gyroscopes. The proposed method was evaluated across several aspects, such as pulse detection, waveform morphology analysis, uniqueness of each generated signal, signal quality enhancement, and heart rate estimation. Additionally, the generalization of the proposed method is examined through testing with unseen users, diverse devices, varying sensor locations, and different subject conditions. Although this method has some limitations, including challenges with envelope filtering and dealing with the data from subjects under intense exercise or anesthesia, it provides a foundation for future advancements.Keywords
This publication has 51 references indexed in Scilit:
- Biased Face Recognition Technology Used by Government: A Problem for Liberal DemocracyPhilosophy & Technology, 2021
- Photoplethysmography based atrial fibrillation detection: a reviewnpj Digital Medicine, 2020
- HeartPy: A novel heart rate algorithm for the analysis of noisy signalsTransportation Research Part F: Traffic Psychology and Behaviour, 2019
- GANomaly: Semi-supervised Anomaly Detection via Adversarial Training2019
- Motion artefact minimization from photoplethysmography based non-invasive hemoglobin sensor based on an envelope filtering algorithmMeasurement, 2018
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Fast Normalized Cross-CorrelationCircuits, Systems, and Signal Processing, 2009
- Framewise phoneme classification with bidirectional LSTM and other neural network architecturesNeural Networks, 2005
- Singular-spectrum analysis: A toolkit for short, noisy chaotic signalsPhysica D: Nonlinear Phenomena, 1992
- A photoelectric plethysmograph for the measurement of cutaneous blood flowPhysics in Medicine & Biology, 1974