Automatic feature selection using wearable sensor data achieved 89% classification accuracy for physical and mental load activities using a custom decision tree or k-nearest neighbor classifier.
Observational (n=12)
Can wearable sensor data and automatic feature selection accurately classify physical and mental load in healthy volunteers?
Wearable sensor data, particularly normalized heart rate, can accurately classify different states of physical and mental load using machine learning algorithms.
Long-term monitoring of health is essential in many chronic conditions, but automatic monitoring is not yet utilized routinely with mental stress. Accelerometers, magnetometers, ECG, respiratory effort, skin temperature and pulse oximetry were used with 12 health volunteers in this study for monitoring 1) heavy mental load, 2) normal mental load, 3) walking, 4) running and 5) lying. Heavy mental load consisted of a 20-min IQ test and normal mental load was represented by reading a comic book. Automatic feature selection using sequential forward search was used to select the best features for classification of the five activities. Normalized heart rate, utilizing activity context information was found to be the most powerful feature for discriminating heavy mental load from normal. Classification accuracy for all 5 activities was 89% with a custom decision tree and with a k-nearest neighbor classifier and 85% with an artificial neural network.
Pärkkä et al. (Mon,) conducted a observational in Physical and mental load (n=12). Wearable sensors (accelerometers, magnetometers, ECG, respiratory effort, skin temperature, pulse oximetry) was evaluated on Classification accuracy for 5 activities (heavy mental load, normal mental load, walking, running, lying). Automatic feature selection using wearable sensor data achieved 89% classification accuracy for physical and mental load activities using a custom decision tree or k-nearest neighbor classifier.