Conventional medical image compression systems employ full-frame processing to avoid the blocking effect, which may lead to the incorrect diagnosis. But full-frame processing is time consuming and, worse, results in the blurring of sharp regions when the compression ratio is high. A new medical image compression technique based on 2 x 2 discrete cosine transform (DCT) is proposed. The 2 x 2 DCT has the ability to retain the detail characteristics of an image. Furthermore, it has the benefit of low computational complexity. By means of a simple transformation to gather the spectrum data with the same frequency domain and an information dominance strategy, the new compression technique promotes efficient coding when applied in the practical data management systems. The simulations are carried out on different types of medical images to evaluate the performance of the proposed scheme. The results verify the efficiency in terms of compression ratio, reconstruction quality, and processing time.