Lossless compression of medical image by hierarchical sorting

We propose a new lossless compression method for medical images, based on a hierarchical sorting. Hierarchical sorting is a technique that achieves a high compression ratio by detecting the regions where image patterns change abruptly, and by sorting pixel order by value to increase predictability. This method enables control of sorting accuracy along with size and complexity. As a result, we can reduce the sizes of the permutation tables and reuse the tables for other image regions. Comparison of this method through experiment reveals better performance for medical images generated by X-ray CT, MRI and large size CR·DR instruments. This technique applies a quad-tree division method to divide an image into blocks in order to support progressive decoding and fast preview of large images.