Wearable sensor-based rehabilitation exercise assessment platform for use in stroke rehabilitation

We are proposing to use wearable sensor based assessment systems to assist the doctors in monitoring their post-stroke patients in the outpatient rehabilitation stage. An important part of developing a performance assessment algorithm is to have a high-precision activity recognition. Convolutional Neural Networks (CNN) are known to give very accurate results, but they require the data to be of a specific structure that differs from the sequential time series format collected from the wearable sensors.