ABSTRACT
Liver cancer is one of the most severe types ofcancerous diseases which is responsible for the death of many patients. Liver cancer images have more noises which is difficult to diagnose the level of the tumor. It is a challenging task to automatically identify the tumor from images because of several anatomical changes in different patients. The cancer is difficult to find because of the presence of objects with same intensity level. In this proposed system, fully automated machine learning is used to detect the liver tumor from input image. Region growing technique is used to segment the region of interest. The textural feature are extracted from Gray level co-occurrence matrix (GLCM) of the segmented image. Extracted textural features are given as input to the designed SVM classifier system. Performance analysis of SVM classification of liver cancer image is studied. This will be useful for physician in better automatic diagnosis of liver caner from input images.
PROJECT VIDEO
.png)
.png)
