Near Lossless Compression at Regions of Medical Interest Using Fuzzy Logic in Designing the Code Book

CT scans are an essential set of medical images. Saving and retrieving these slices are of high value. Therefore, with regard to the huge volume of data and restrictions in data storage media or the network’s band width, employing compression techniques seem to be quite vital. The method suggested in this article for the compression of liver CT scans is the first stage in image segmenting and defining the regions of interest. The whole image is then compressed through lossy compression method using VQ. To design the code book in this algorithm, the fuzzy logic and C-means clustering methods are recommended. The erroneous image is finally computed within the region of interest. The results ultimately indicate the superiority of the presented algorithm over LZ, JPEG2000 and JPEG-LS compression standards.

Keywords:

C-means clustering , designing the code book , fuzzy logic , image compression