Muhammad Anas Javiad2015-05-192015-05-192015https://escholar.umt.edu.pk/handle/123456789/1548The rapid development in medical imaging provided us new knowledge about diseases and complex structure of human body. New technology is being developed to help improve global healthcare system. The aim of this research is to develop an efficient algorithm to compress grayscale medical images at the cost of speed and memory. We have developed lossless, lossy and hybrid image compression algorithms. A group of MRI, X-Ray, CT-scan, Ultra sound, Angiograms and Mammograms images were used for experiments. The performance of purposed algorithm was compared with a state of art JPEG2000 compression technique. The lossy compression technique follows some JPEG base line standard. The overall compression ratio is 2 times better than JPEG with same PSNR. The lossless technique shows 1.2 times better compression ratio than PNG. In medical images only a small portion is diagnostically important hence in ROI diagnostically important area is compressed lossless and remaining area is compressed lossy. Hybrid compression is based on Region of Interest (ROI) coding technique. The overall compression ratio is 4 times better than JPEG and 9 times better than JPEG2000 with same PSNR. Proposed techniques perform well regarding to the compression ratio and quality of reconstructed image, making them suitable for medical purposes.BS ThesisMedical Image CompressionThesis