Unsupervised Classification of Liver Lesions Using Automated Image Segmentation and Classification Model

M. Shenbagapriya and Dr.M. Vanitha

In this paper, we propose an Automated Diagnosis Model (ADM) to group enormous arrangements of liver injuries. In this model, we at first pre-process the picture to dispose of the clamors present in the examples. Furthermore, we section the picture into names utilizing improved Gray Wolf Optimization (GWO) with edge-based division. At last, the names are arranged into separate malady utilizing Support Vector Machine (RNN-SVM) classifier, individually. This structure is prepared for grouping the specific injury over the huge assortment of liver sore picture database and afterward it is tried on ongoing pictures. The test consequences of ADM is contrasted and existing techniques as far as precision, affectability and particularity. The ADM with improved division and grouping recognizes the classes of liver injuries. The exploratory outcomes show that ADM gets preferred characterization over regular classifiers as far as precision, explicitness and affectability.

Volume 12 | Issue 8

Pages: 219-225

DOI: 10.5373/JARDCS/V12I8/20202467