In many biomedical image processing, image features plays predominant role to discriminate the abnormality from normal image classes. And for improved exploration of various object of interests input image enhancement is also widely preferred and considered as potential pre processing block in many CAD system but this will causes some significant changes in basic image characteristics with associated clinical relevance. These non linear changes while adjusting the brightness causes image artifacts as well in the output enhancement image. To mitigate this issues, in this paper grey wolf optimization algorithm based Histogram Equalization is introduced to formulate the threshold limits which can control the saturations of output pixels and avoids over enhancements. Hardware implementation of proposed GWO based HE model is validated with its improved image feature retention over output enhanced image. Finally to narrow down the complexity overhead in HE process with the inclusion of complex GWO optimization model high performance RNS arithmetic and carry free parallel prefix accumulator designs are incorporated which showed significant complexity and path delay reduction with improved energy efficiency.