Automatic Cell Nucleus Segmentation Using Superpixels and Clustering Methods in Histopathological Images

It is seen that there an increase in cancer and cancer-related deaths day by day. Early diagnosis vital for the early treatment of cancerous area. Computer-aided programs allow unhealthy cells specialist pathologists diagnose as a result efforts. In this study, kMeans Fuzzy C Means methods, which are among global segmentation SLIC, Quickshift, Felzenszwalb, Watershed ERS algorithms, superpixel were used automatic cell nucleus detection high resolution histopathological images with computer aided programs. As success performances algorithms analyzed evaluated. better obtained watershed FCM used. Quickshift SLIC methods gave results terms precision. k-Means provide best performance F measure (F-M) true negative rate (TNR) more successful methods.