Medical Image Compression Algorithm Using Edge-Aware Adaptive Block Partitioning

Abstract

: As the world witnesses a major advance in medical staff diagnosis depending on medical images, including MRI and CT scans, it requires having compression approaches that balance between preserving essential medical information and reducing medical image data size. Such approaches that depend on local variance for the adaptive block partitioning mechanism , which is distinct form the classic method, like JPEG , in how usually experience blocking artifacts and loss of fine details. Instead of utilizing explicit edge detection, the algorithm uses both integer wavelet transform (IWT) and adaptive bit allocation to differentiate between smooth non-structural areas and structural regions containing vital edges. The algorithm shows superiority through applying the proposed algorithm on several medical images by achieving a Structural Similarity Index (SSIM) consistently exceeding 0.996 and an average Peak Signal-to-Noise Ratio (PSNR) of 48.01Db, which overcomes the performance of the JPEG standard in structural fidelity with the accomplishment of as equiponderant compression ratio (averaging 1.58) . The proposed algorithm is ideal for clinical application where accurate reconstruction in crucial, keeping diagnostically important structures and preserving tissue outlines.