The High-Resolution Medical Picture Compression Using Segmentation-Based Lossless Image Coding

Medical imaging serves a crucial role in the diagnosis of illness. The amount of local storage capacity and transfer bandwidth needed by remote medical devices to aid in patient diagnosis and treatment is closely related to the losslessness of the compression used. Losslessness and similarity are two amazing features of medical imaging. Compression centres on learning to take use of these two features to decrease the amount of data that must be stored to describe a picture. In this study, we use large dataset mining to create an image catalogue. That is, to identify pictures' elemental parts. For multi-component medical pictures, we offer a soft compression approach that faithfully preserves the images' underlying structure. Our findings demonstrate that our newly developed soft compression approach outperforms the industry standards JPEG2000, and PNG and we accompany this with thorough representation architecture for image compression.