DeepHeart, a novel deep-learning-based denoising approach, achieved an average absolute error of 1.61 bpm for heart rate estimation, outperforming classical signal processing (2.34 bpm).
Does DeepHeart improve the accuracy of heart rate estimation from contaminated PPG signals compared to existing approaches?
DeepHeart provides a highly accurate method for estimating heart rate from motion-contaminated PPG signals, outperforming existing deep learning and classical signal processing methods.
Absolute Event Rate: 1.61% vs 2.34%
Heart rate (HR) estimation based on photoplethysmography (PPG) signals has been widely adopted in wrist-worn devices. However, the motion artifacts caused by the user’s physical activities make it difficult to get the accurate HR estimation from contaminated PPG signals. Although many signal processing methods have been proposed to address this challenge, they are often highly optimized for specific scenarios, making them impractical in real-world settings where a user may perform a wide range of physical activities. In this article, we propose DeepHeart, a new HR estimation approach that features deep-learning-based denoising and spectrum-analysis-based calibration. DeepHeart generates clean PPG signals from electrocardiogram signals based on a training data set. Then a set of denoising convolutional neural networks (DCNNs) are trained with the contaminated PPG signals and their corresponding clean PPG signals. Contaminated PPG signals are then denoised by an ensemble of DCNNs and a spectrum-analysis-based calibration is performed to estimate the final HR. We evaluate DeepHeart on the IEEE Signal Processing Cup training data set with 12 records collected during various physical activities. DeepHeart achieves an average absolute error of 1.61 beats per minute (bpm), outperforming a state-of-the-art deep learning approach (4 bpm) and a classical signal processing approach (2.34 bpm).
Chang et al. (Sat,) conducted a other in Heart rate estimation from contaminated PPG signals (n=12). DeepHeart (deep-learning-based denoising and spectrum-analysis-based calibration) vs. State-of-the-art deep learning approach and classical signal processing approach was evaluated on Average absolute error in beats per minute (bpm). DeepHeart, a novel deep-learning-based denoising approach, achieved an average absolute error of 1.61 bpm for heart rate estimation, outperforming classical signal processing (2.34 bpm).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: