A neural network to diagnose liver cancer

Author

Maclin, Philip S. ; Dempsey, Jack

Author_Institution

Tennessee Univ., Memphis, TN, USA

fYear

1993

fDate

1993

Firstpage

1492

Abstract

A backpropagation neural network is designed to diagnose five classifications of hepatic masses: metastatic carcinoma, hepatoma (HCC), cavernous hemangioma, abscess, and cirrhosis. BrainMaker Professional version 2.5 software is used in this research. The input submitted to the network consists of 35 numbers per patient case, which represents ultrasonographic data and laboratory tests. The network architecture has 35 elements in the input layer, two hidden layers of 35 elements each, and five elements in the output layer. After being trained to a learning tolerance of 1%, the network classifies hepatic masses correctly in 51 of 72 cases. Continued research should provide a computerized second opinion that will be especially helpful to clinicians

Keywords

backpropagation; learning (artificial intelligence); medical diagnostic computing; neural nets; BrainMaker Professional version 2.5; abscess; backpropagation; cavernous hemangioma; cirrhosis; hepatic masses; hepatoma; learning; liver cancer diagnosis; medical diagnostic computing; metastatic carcinoma; neural network; Abdomen; Artificial neural networks; Biological neural networks; Cancer; Laboratories; Liver; Magnetic resonance imaging; Metastasis; Neural networks; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1993., IEEE International Conference on

Conference_Location

San Francisco, CA

Print_ISBN

0-7803-0999-5

Type

conf

DOI

10.1109/ICNN.1993.298777

Filename

298777