mmCare: A Nursing Care Activity Monitoring System via mmWave Sensing
Peilin Pan, Fenglin Zhang, Anfu Zhou, Huadóng Ma, Huaizhen Jia · 2024
With the advent of an aging society, an increasing number of patients require appropriate care. Simultaneously, there is currently a lack of effective solutions for monitoring nursing behaviors. In order to address this issue, we propose mmCare, a nursing care activity detection system based on mmWave sensing, capable of achieving comparatively accurate and effective monitoring while preserving privacy. Through standardization processing of point cloud information into point cloud pictures and extraction of spatio-temporal features, mmCare ultimately identifies 7 categories of nursing actions. Through practical deployment and evaluation of real-world datasets, mmCare achieves an accuracy rate of 93.52% on the individual test dataset.