Lossy compression in nuclear medicine images.

Abstract

The goal of image compression is to reduce the amount of data needed to represent images. In medical applications, it is not desirable to lose any information and thus lossless compression methods are often used. However, medical imaging systems have intrinsic noise associated to it. The application of a lossy technique, which acts as a low pass filter, reduces the amount of data at a higher rate without any noticeable loss in the information contained in the images. We have compressed images of nuclear medicine using the discrete cosine transform algorithm. The decompressed images were considered reliable for visual inspection. Furthermore, a parameter was computed from these images and no discernible change was found from the results obtained using the original uncompressed images.