Date of Publication :31st January 2022

Abstract: Initial prediction of any kind of cancer is always advantageous for on-time medical treatment to save the patient's life. The Computer-Aided Diagnosis (CAD) tools using signal processing & image processing methods gained significant attention for immediate & accurate diagnosis using patient’s raw medical data like Magnetic Resonance Imaging (MRI), Chromatography (CT), etc. The liver cancer early detection & analysis of its grading is an important research problem. In this research, we proposed the two models semi-automatic & automatic frameworks for liver disease classification. The models perform early detection of liver cancer accurately followed by its grading analysis into different stages like stage 1 (T1), stage 2 (T2), & stage 3 (T3). The proposed framework consists of stages like pre-processing, Region of Interest (ROI) extraction, features extraction, & classification. The raw CT scans of the liver are pre-processed to remove the noises using the filtering & contrast adjustment functions. The adaptive segmentation method is designed to using binarization & morphological operations to extract the accurate ROI with the minimum computational burden. For features extraction, the text features extracted using Gray Level Co-occurrence Matrix (GLCM), shape features using geometric moment, & automatic features using Convolutional Neural Network (CNN). The hybrid form of features normalized using the min-max technique. For the classifications, we explored the classifiers such as Artificial Neural Network (ANN), Support Vector Machine (SVM), & Long Term Short Memory (LSTM). We investigated the semi-automated & automated systems using the publically available research dataset

Reference :