A machine learning approach to control a Prosthetic arm via signals from residual limb - A boon for amputees

Many people have lost their hand due to accidents. This leads to upper or lower arm amputation, wrist disarticulation, trans-radial amputation, commonly referred to as below-elbow or forearm amputation. People who lost their hand face a lot of problems physically, socially and also undergo mental traumas. They were also seen as people with disabilities which makes them feel even more inferior.Hence, this work aims to develop a prosthetic limb or arm for individuals who have lost their hand due to accidents. Such individuals face numerous physical, social, and mental challenges and are often perceived as disabled. To address this issue, this work proposes using s-EMG signals acquired from residual arm muscles to develop an artificial hand that can be controlled through muscle movements. To achieve this, the sEMG signal needs to be classified for different actions, as each muscle movement corresponds to a distinct pattern in the signal domain. The ultimate goal of this work is to provide individuals who have lost their hand with a prosthetic arm that functions as a natural hand and improves their quality of life.