We present two types of sensor data analysis for medical and healthcare. One sensor dataset is collected in a hospital for medical purposes. We gathered accelerometer data and RFID data of real nursing in the hospital. We provide the real nursing dataset for mobile activity recognition which could be used for supervised machine learning, and also the big data combined with the patients’ medical records and sensors tried for 2 years. The other sensor dataset is collected in a developing country. We developed an eHealth system that comprises a set of sensor devices in an attache case. The first checkup was provided to 16,741 subjects. After 1 year, 2361 subjects participated in the second checkup, and the blood pressure of these subjects was significantly decreased (P<0. 001). Based on these results we proposed a cost-effective method using a predictor, to ensure sustainability of the program in developing countries.
!!!All Science Journal Classification (ASJC) codes
- コンピュータ サイエンス（全般）