Development of an embedded EMG-based wristband for hand gesture recognition using machine learning algorithms.

With the recent improvement of flexible electronics, wearable devices are becoming more and more non-invasive and comfortable, pervading fitness and health-care applications. Wearable devices allow unobtrusive monitoring of vital signs and physiological parameters, enabling advanced Human Machine Interaction (HMI) as well. On the other hand, battery lifetime remains a challenge especially when they are equipped with bio-medical sensors and not used as simple data logger. In this thesis, we present a flexible wristband, designed on a flexible PCB strip, for real-time EMG-based hand gesture recognition. Experimental results show the accuracy achieved by the algorithm and the system implementation. The proposed wristband executes a Support Vector Machine (SVM) algorithm reaching up to 96% accuracy in recognition of 5 hand gestures collecting data from 5 users. The system targets health-care and HMI applications, and can be employed to monitor patients during rehabilitation from neural traumas as well as to enable a simple gesture control interface (e.g. for smart-watches).