The technique which is most important for the Reversible Data Hiding (RDH) is Prediction Error Expansion (PEE) through this we can hide large messages into digital media with little distortion. The Nearest Neighborhood Pixel Prediction (NNP2) algorithm which goes under the process of the Rhombus prediction and Chinese Remainder Theorem (CRT). In which these several methods are used for the prediction errors, variation of prediction and pixel are used to embed data. The combined Chinese Remainder Theorem as well as Rhombus prediction provides better quality output rather than traditional one. For embedding several bits into one embeddable pixel the size have to be modified and adjusted using Chinese Remainder Theorem (CRT). The distortion level is reduced by Histogram Shifting (HS). The performance of proposed method is evaluated using PSNR for number of medical images. The simulation results provide the best encoded capacity as well as good image pattern than ancient techniques.