One of the most important reasons of heart attack is the narrowing of arteries. Coronary angiography is an efficient x-ray examination for diagnosing coronary artery diseases and diagnosing blocked arteries. Also the abnormalities in the blood vessels such as stenosis can be diagnosed. During the angiography procedure, a narrow tube, that is called a catheter, enters the vessel from groin region. After placing catheter in the heart or in the entrance of the feeding arteries of the heart a contrast agent is injected through. Then a number of X-Ray images, which are called angiogram, are taken. Blood vessels become visible after the injection and an expert can diagnose coronary artery diseases without the need for a hard invasive operation . For a patient this procedure takes 4 to 5 minutes during which about 4000 images are taken. The size of these images is typically 512×512 pixels. Usually these images are taken at a speed of 30 frames per second with at least 8-bit pixels. The size of the raw data in this procedure is approximately 2.5 GB. torage and transmission of this large amount of data is a challenge for hospitals. Efficient compression of these data files is an inevitable requirement. Image sequences from digital angiography contain areas of high diagnostic interest. Loss of information due to compression for regions of interest (ROI) in angiograms is not tolerable. Due to the sensitivity of the medical data in medical images, lossless methods are preferred. But the maximum compression ratio in these methods is about 2 or 3:1. The compression ratio in the Lossy methods is always higher than those of the lossless methods. On the other hand, in the lossy methods some important medical details may be lost and diagnostic errors may occur . By analysing angiogram images it is concluded that large areas of each frame does not contain diagnostically important information. Therefore it is concluded that using of lossy methods for some parts of an image can be medically acceptable. In the literature a lot of work is done on the lossless and lossy compression of the angiogram images. Some of these researches identify a region of interest (ROI) and allocate more bit budget for the diagnostically important regions. These methods are based on both 2D and 3D techniques similar to the video compression methods. Key Words Coronary Angiography, angiogram sequences, Region of Interest, compression, context modeling