A New Diagnosis Lossless Compression Method for Digital Mammography

With the rapidly growing use of digital images in medical archival and communication, image compression technology, especially diagnosis lossless compression technology, plays a more and more important role for medical applications. In this thesis, a novel diagnosis loseless compression algorithm is presented for digital mammography. The mammogram is divided into breast region, pectoral muscle and background using the CAD technologies. Then the breast region and pectoral muscle are compressed losslessly and lossily using arbitrary shape ROI compression algorithm respectively. The background can be discarded or compressed as user´s will. Experimental results show that the proposed method offer potential advantage in medical applications of digital mammography compression.