Our human fingers are provided specific roles which finally results into various hand motions and functions. The thumb plays a special role because it takes part in several activities. So, a thumb loss because of traumatic accidents or other reasons can prove terrible because appropriate functions of the hand will get sternly restricted. Prosthetic thumb is developed, for solving this problem, and to be worn to counterpart functions of the lost thumb. The control strategy of the prosthetic part is completely based on the surface electromyogram (sEMG) signals to control it. This paper uses the electrical activity of the muscles (EMG) as control signals in a pattern recognition control system. The paper proposes a system for pattern recognition of EMG signals using MATLAB and ANN. © 2018 IEEE.