[B-19] Healthcare and Medical Information Communication Technology

11:00 AM - 11:15 AM JST(2:00 AM - 2:15 AM UTC)Candidate for Academic Incentive AwardCareer Explorer

[B-19-06] Non-Contact Heart Rate Estimation Using a 60 GHz Pulsed Radar and 2D
ResNet: A Study on Training Data Optimization and Distance Dependency

◎△yuuki nagai¹, Taiyo Takahashi², Takeshi Toda¹ (1. College of Science & Techbology, NIHON UNIVERSITY, 2. Graduate School of Science & Techbology, NIHON UNIVERSITY)

Keywords:

Heart rate,Healthcare,Pulse sensor,Radar,Vital signs

Conventionally, FMCW radar has been the dominant technology for non-contact vital sign detection. In our previous work, we investigated heartbeat estimation using deep learning — specifically CNN and 2D ResNet — applied to a low-power pulse-based sensor, achieving MAEs of 9.54 BPM and 9.03 BPM, respectively. However, the subject composition of the training data and the measurement distances were limited, leaving the evaluation under more practical conditions as a remaining challenge. In this study, we revisited the training data (subject data) and conducted evaluations at relatively longer distances, examining the generalization performance of the model and its robustness to variations in distance.