Deep Learning Methods for Remote Heart Rate Measurement: A Review and Future Research Agenda

Abstract

Heart rate (HR) is one of the essential vital signs used to indicate the physiological health of the human body. While traditional HR monitors usually require contact with skin, remote photoplethysmography (rPPG) enables contactless HR monitoring by capturing subtle light changes of skin through a video camera. Given the vast potential of this technology in the future of digital healthcare, remote monitoring of physiological signals has gained significant traction in the research community. In recent years, the success of deep learning (DL) methods for image and video analysis has inspired researchers to apply such techniques to various parts of the remote physiological signal extraction pipeline. In this paper, we discuss several recent advances of DL-based methods specifically for remote HR measurement, categorizing them based on model architecture and application. We further detail relevant real-world applications of remote physiological monitoring and summarize various common resources used to accelerate related research progress. Lastly, we analyze the implications of research findings and discuss research gaps to guide future explorations.

Stress Management Intervention for Leaders Increases Nighttime SDANN: Results from a Randomized Controlled Trial

“…HRV not only represents a physiological parameter correlated with both stress and disease, but it also has properties that facilitate wider application in that the required RR interval measurement is non-invasive and can be obtained at low cost. With advances in digital health technologies, techniques using deep learning, artificial intelligence, and neural networks are emerging [ 35 , 36 , 37 , 38 ], some of which can approximate heart rate without the need for a device to be worn directly on the skin [ 39 , 40 ]. This rapidly developing area holds the potential for widespread application, especially for the field of preventive medicine in the workplace.…”

Heart rate prediction from facial video with masks using eye location and corrected by convolutional neural networks

“…We can clearly observe that the results of all evaluation metrics decreased significantly. A mean reason for this phenomenon is the videos are heavily compressed, so noise artifact was unavoidably added [34] , [55] . McDuff et al [56] also find that a considerable drop in SNR between raw and compressed videos through experiments.…”

“…Besides, using deep learning (DL) approaches in the rPPG field have emerged and yielded excellent results in recent years. Ni et al [33] and Cheng et al [34] provided a comprehensive review of DL-based methods. Since training network needs a large amount of data collected in various real scenes, ensures the robustness and flexibility of the DL method for practical application.…”

Diagnostic Features and Potential Applications of PPG Signal in Healthcare: A Systematic Review

“…iPPG discovery opened up a new field of research and played an important role in remote patient monitoring experiments [ 14 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 ]. Deep learning hype is also brought to the field of iPPG [ 14 , 15 ].…”

“…Several reviews on the application of PPG signals have been carried out. Some of them focus on the specific medical use of PPG like pulse rate [ 3 ], blood pressure [ 4 , 5 ], atrial fibrillation [ 6 , 7 ], circulatory monitoring [ 8 ], nociception [ 9 ], or on the specific placement of PPG [ 10 , 11 ], others instead focused on reviewing the way that the signal has been analyzed [ 7 , 12 , 13 , 14 , 15 , 16 ] or the type of the sensor [ 17 ]. More than a decade ago, J. Allen [ 2 ] published an interesting review about the applications PPG in clinical physiological measurement.…”