Perceptually lossless image coding of medical images with artificial edge segmentation

This paper presents a method of encoding medical images with artificial edges at a perceptually lossless quality. Based on the JPEG2000 coding framework and embedded with an advanced perceptual distortion metric, the proposed coder can identify and remove visually insignificant/irrelevant information. In addition, an artificial edge segmentation algorithm complements the perceptual distortion metric by enabling it to adaptively identify and prune artificial edges. Compared with the state-of-the-art JPEG compliant LOCO lossless coder, current results have shown an improvement in bit-rate without any loss of visual fidelity.