Computational Analysis and Deep Learning for Medical Care

9Computer-Aided Diagnosis of Liver Fibrosis in Hepatitis Patients Using Convolutional Neural Network

Aswathy S. U.^1*, Ajesh F.², Shermin Shamsudheen³ and Jarin T.⁴

¹Department of Computer Science and Engineering, Jyothi Engineering College, Thrissur, Kerala, India

²Department of Computer Science and Engineering, Anna University, Chennai, Tamilnadu, India

³Faculty of Computer Science & Information Technology Jazan University, Jazan, Saudi Arabia

⁴Department of Electrical and Electronics Engineering, Jyothi Engineering College, Thrissur, Kerala, India

Abstract

Diagnosis of diseases like liver fibrosis is one of the quintessential part in medical areas. With the help of historical data of patient’s, the respective need is to make decision for further process. To achieve a greater accuracy and timely decision is always complex due to its dynamic nature, blurriness, and uncertainty associated with that disease. This paper gives the solution for the above-mentioned problem with diagnosis of liver patients. This objective study takes liver image sets over five categories (category A, classic hepatocellular carcinomas [HCCs]; category B, malignant liver tumors; category C, indeterminate masses or mass-like lesions and rare benign liver masses; category D, hemangiomas; and category E, cysts). The proposed CNN model is VGG-16 inspired SegNet which is composed of 13 convolutional layers, three fully connected layers in a encoder-decoder network. This was tested with more than 100 liver ...