Contactless Heart Rate Variability (HRV) Estimation Using a Smartphone During Respiratory Maneuvers and Body Movement

DSE-NN: Deeply Supervised Efficient Neural Network for Real-Time Remote Photoplethysmography

Non-contact remote photoplethysmography can be used in a variety of medical and healthcare fields by measuring vital signs continuously and unobtrusively. Recently, end-to-end deep learning methods have been proposed to replace the existing handcrafted features. However, since the existing deep learning methods are known as black box models, the problem of interpretability has been raised, and the same problem exists in the remote photoplethysmography (rPPG) network. In this study, we propose a method to visualize temporal and spectral representations for hidden layers, deeply supervise the spectral representation of intermediate layers through the depth of networks and optimize it for a lightweight model. The optimized network improves performance and enables fast training and inference times. The proposed spectral deep supervision helps to achieve not only high performance but also fast convergence speed through the regularization of the intermediate layers. The effect of the proposed methods was confirmed through a thorough ablation study on public datasets. As a result, similar or outperforming results were obtained in comparison to state-of-the-art models. In particular, our model achieved an RMSE of 1 bpm on the PURE dataset, demonstrating its high accuracy. Moreover, it excelled on the V4V dataset with an impressive RMSE of 6.65 bpm, outperforming other methods. We observe that our model began converging from the very first epoch, a significant improvement over other models in terms of learning efficiency. Our approach is expected to be generally applicable to models that learn spectral domain information as well as to the applications of regression that require the representations of periodicity.

Camera-based Cardiovascular Screening based on Heart Rate and Its Variability In Pre- and Post-Exercise Conditions

Analysis of heart rate (HR) and heart rate variability (HRV) in pre- and post-exercise conditions can provide useful information about the health condition of cardiovascular system. Remote photoplethysmography (rPPG) that uses a contactless camera to measure vital signs from the human face might allow for ubiquitous applications in the fitness scenario. This paper benchmarks the accuracy of HR and HRV measured by ECG, PPG and rPPG, and investigates the difference of HR and HRV parameters between individuals with/without the exercise habit in pre- and post-exercise phases. We built a fitness benchmark consisting of 14 healthy subjects to perform running exercise on a treadmill, with video and reference data recorded simultaneously. The results show that rPPG has a similar performance as PPG in both the estimation of HR and HRV. The HRV parameters (Mean IBI, SDNN, LF, VLF, and SD2) of rPPG/PPG show good agreements with the ECG reference. Subjects with a regular exercise habit show lower resting HR and smaller values of HRV parameters in pre-exercise phase, fewer changes in HR and HRV at the same exercise intensity, and faster HR recovery after exercising, as compared with those without an exercise habit. The preliminary results suggest that rPPG is a promising surrogate of PPG for screening HR and HRV before and after exercise, to indicate the performance of cardiac training.