Machine Intelligence, Big Data Analytics, and IoT in Image Processing

3Liver Cancer Classification With Using Gray-Level Co-Occurrence Matrix Using Deep Learning Techniques

Debnath Bhattacharyya¹*, E. Stephen Neal Joshua² and N. Thirupathi Rao²

¹ Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India

² Department of Computer Science and Engineering, Vignan’s Institute of Information Technology, Visakhapatnam, AP, India

Abstract

Liver cancer is the most common cause of death in the world. In order to find liver cancer, you need to figure out what medical images mean adaptive thresholding with a watershed transform is used to make the liver stand out from the other parts of the body. Optimal strategies and the swarm optimization model are some of the methods used to separate the malignant area of the liver. The data included 225 images from patients with different types of liver cancer. In order to build a big dataset, the Gray Level Matrix, the Local Binary Pattern was used to find features that are important. It then gets broken down into different types of cancer using neural network, support vector machine, random forest, and deep neural network classifiers. These include hemangioma, hepatocellular carcinoma. The suggested methods are judged on their sensitivity, specificity, accuracy, and Jaccard index. People who use watershed Gaussian-based deep learning algorithms have found that they can be used to diagnose liver cancer. DNN classifiers were found to be the best, ...