Medical Imaging Classification Of Compression Methods

With the rapid development of internet technology and the research of image compressing method, new medical treatment is produced, such as long-distance medical diagnosis, long-distance medical education, long-distance operation, and so on. Patients can gain consultation and cure of long-distance experts, which can save time and money of doctors and patients. At the same time, storage and transmission of medical image come to be the key. Most of long-distance medical systems are realized in high-speed networks, and internet can not realize this speed. Therefore, for the rapid transmission of medical images on the Internet, we have to compress them.In principle, for the need of diagnosis, medical images are often required a high precision, so digital medical images require large amount of storage. In order to ensure their quality, medical images are compressed losslessly when transmission. In fact, medical imaging data is much greater than ordinary image, therefore, these images occupy a large of bandwidth, and also reduce the transmission rate. For these reasons, lossless compression efficiency is too low which makes it difficult for a real medical imaging compression. In order to make use of network bandwidth and storage space, lossy compression is very important.The purpose of my study is to enhance the quality of medical images affter lossy compression, and to guarantee the authenticity of diagnostic medical images by the lossy compression. For diagnosis, focus regions are the regions they interested, but the non-focus regions usually hold great information. Diagnosis is not interested in these regions, so if these informations can be reduced, the bandwidth will be able to effectively reduced. On the other hand, some focus regions of medical images are not clear, this brings difficulty for the diagnosis, therefore, we must enhance the medical image.This article proposes one compressing method which feels the interested region and the non-interested region to realize the medical image transmission. We combine lossy compression and lossless compression. With the actual need, we will adopt the different compression rate to each region, and use JPEG2000 standard to carry on the multi-scale compression. For the region of interest (focus region), we carry on lossless compression, but for the non-interest region (non-focus region), we use lossy compression. This can guarantee the important information has the better visual effect, and not affect the diagnosis result, which causes the entire image to satisfy the transmiting request. This article completed below work:(1) Carried on the medical image pre-processing—image enhancement. This process ing enabled the contrast of primitive image to obtain increases, and made it clearlier. This article improved the classics enhancement method-partial edge examination method, and unifies the person eye visual characteristic.(2) Carried on the division of medical images. This article gained the region of interest (focus region) using the division method. It used the threshold division method, and obviously separated the interest region and the non-interest region, which is favor to the multi-scal compression.(3) Compressed the image with different scale (namely different compression rate). It used JPEG2000 standard to carry on the compression. For focus region, it carried on lossless compression, and to the non-focus region, used lossy compression. This can guarantee the important information has better visual effect, and not affect the diagnosis result, which causes the entire image to satisfy the transmission request.This article used the matlab language to carry on the realization of the algorithm. It finally indicated: using this method can obtain good visual effect, and gained greatly effectively to the compression rate, thus improved the transmissing speed.