PERFORMANCE ANALYSIS OF DISTANCE MEASURES FOR MEDICAL IMAGE SEGMENTATION

Method proposed use the k-means clustering algorithm with hybrid mathematical law derived from the laws of former scientists that extracts the least distance to segmentation for medical image containing the disease . Proposed method evaluation is based on the comparison with the results of skin cancer images in previous researches using performance criteria sensitivity , specificity ,accuracy .in our research we obtained the following criteria .accuracy 94.28%,specificity 93.75%,sensitivity 100% with 2 false diagnosis of the 35 samples