Using K-mean Clustering to Classify the Kidney Images

This study has applied digital image processing on three-dimensional C.T. images to detect and diagnose kidney diseases. Medical of different cases diseases were compared with those healthy cases. Four kidneys disorders, such as stones, tumors (cancer), cysts, renal fibrosis considered in additional tissues. method helps differentiating between the diseased It can its very early stages, before they grow large enough be seen by human eye. The used for segmentation texture analysis was k-means co-occurrence matrix. separates classes tumor classes, affected parts isolated from parts. To isolate other anatomical a CT image, mask must generated, which is binary (0s or 1s). also utilized remove undesired characteristics images. Density slicing color based density. A slice band neighboring gray levels scale through monocular color. (0-255) transformed into variety slices; it conversion colored that efficiently displays symmetric diverse regions. property process segmentation. unsupervised classification process, K-Mean clustering, application K-mean classify type kidney. clustering each class depending properties distance separate part tissue; Co-occurrence matrices gives statistical energy, homogeneity, contrast, correlation. These give an indication nature tissues are standard deviation cancer higher than stone, so mean, contrast means brighter none grey level more stone this energy value; highly correlated. proved good diagnosis.