Wearable Sensor Data for Activity and Health Monitoring

The independent variables considered come from the raw data of wearable sensors. Previous works used accelerometer, gyroscope and heart rate sensors for monitoring physical activity, health conditions, cognitive, social and other factors [49 (link),54 (link),55 (link),56 (link),57 ]. We have used similar sensors built-in a low-cost smartwatch, the Samsung Gear S3 [58 ]. Samsung Gear S3 supports development of software applications on its open access Tizen operating system. This characteristic allows to collect the sensory raw data and to develop software applications in a flexible manner. The built-in sensors we used are: (1) the triaxial accelerometer, which measures changes in the device velocity; (2) the triaxial gyroscope, which detects angular velocity and orientation of the device; (3) and the heart rate sensor, which measures the heart beats per minute. In particular, we collected the raw data from these wearable sensors obtaining 7 variables, detailed in Table 2. The value of these variables was provided in integer number (without a decimal point), or in float (floating-point number, which is a number that has a decimal place).