A Review of Classifying the Hepatocellular Carcinoma using ML and Image Processing Techniques

Abstract

As a part of this study, an intricate and exhausting review has been done on classify the liver lesion using Deep Learning (DL) and Image Processing Techniques. Despite that, Computerized Axial Tomography (CAT) Scan or CT-Scan,Nuclear Magnetic Resolution Image(NMRI) or (MRI),Ultrasound (U/S),positron emission tomography-computed tomography (PET-CT) scan are most recommended techniques used for cancer testing and diagnosis. However, doctors and radiologists found difficulty in detecting and identifying cancer from these techniques. Since Computer Aideddiagnosis can be crucial for detecting cancerous cells in the body to such an extent, doctors can givethe best treatment to the patient accordingly. Various machine-learning and image-processing computer-aided techniques have been explored, tested, and also accomplished in the medical field. The primary intention of the paper is to examine the diverse Computer – Aided strategies, correlate these techniques with each other and evaluate the best technique with their limitations and shortcoming. Various types of machine learning techniques are used in the healthcare field such as Decision trees, Artificial Neural Network (ANNs), Space Invariant Artificial Neural Networks or ConvNet, Bayesian Networksetc. By using these approaches, it is to be seen that there is a lack of accuracy in classifying the liver lesion in the initial stages.