KID-PPG: Knowledge Informed Deep Learning for Extracting Heart Rate from a Smartwatch

TL;DR

KID-PPG is a novel deep learning approach that integrates expert knowledge to improve heart rate extraction from PPG signals, effectively addressing motion artifacts and signal degradation.

Contribution

This paper introduces KID-PPG, a knowledge-informed deep learning model that combines expert signal processing techniques with probabilistic inference for better heart rate estimation.

Findings

Achieved an average MAE of 2.85 bpm on PPGDalia dataset.

Outperformed existing methods in heart rate tracking accuracy.

Demonstrated the benefit of incorporating prior knowledge into deep learning models.

Abstract

Accurate extraction of heart rate from photoplethysmography (PPG) signals remains challenging due to motion artifacts and signal degradation. Although deep learning methods trained as a data-driven inference problem offer promising solutions, they often underutilize existing knowledge from the medical and signal processing community. In this paper, we address three shortcomings of deep learning models: motion artifact removal, degradation assessment, and physiologically plausible analysis of the PPG signal. We propose KID-PPG, a knowledge-informed deep learning model that integrates expert knowledge through adaptive linear filtering, deep probabilistic inference, and data augmentation. We evaluate KID-PPG on the PPGDalia dataset, achieving an average mean absolute error of 2.85 beats per minute, surpassing existing reproducible methods. Our results demonstrate a significant performance improvement in heart rate tracking through the incorporation of prior knowledge into deep learning models. This approach shows promise in enhancing various biomedical applications by incorporating existing expert knowledge in deep learning models.