Efficient prediction and classification for cirrhosis disease using LBP, GLCM and SVM from MRI images

To enhance the specificity of Magnetic resonance imaging (MRI) based cirrhosis stage-diagnosis, a method of diagnosis incorporating the scan image texture discovery with classification techniques are suggested. First of all the liver MRI images are preprocessed and Region of Interest (ROI) image field is retrieved from the MRI images and the Lloyd method is compressed and quantified. Then a Local Binary Pattern (LBP) process filtering the ROI image and then extracts the texture features of the LBP images from four directions with a twenty directional Gray-Level Co-occurrence Matrix (GLCM). After that, the Support Vector Machine (SVM) technique has been used for MRI image classification and cirrhosis disease prediction. The experimental findings indicate that the procedure suggested will diagnose hepatic cirrhosis correctly.