Neural Network Approach to Characterization of Cirrhotic Parenchymal Echo Patterns

Shin-ya YOSHINO  Akira KOBAYASHI  Takashi YAHAGI  Hiroyuki FUKUDA  Masaaki EBARA  Masao OHTO  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E76-A   No.8   pp.1316-1322
Publication Date: 1993/08/25
Online ISSN: 
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section of Papers Selected from the 7th Digital Signal Processing Symposium)
Category: Biomedical Signal Processing
computer-aided diagnosis,  image texture,  liver cirrhosis,  neural network applications,  ultrasonic images,  

Full Text: PDF(904.8KB)>>
Buy this Article

We have calssified parenchymal echo patterns of cirrhotic liver into four types, according to the size of hypoechoic nodular lesions. Neural network technique has been applied to the characterization of hepatic parenchymal diseases in ultrasonic B-scan texture. We employed a multi-layer feedforward neural network utilizing the back-propagation algorithm. We carried out four kinds of pre-processings for liver parenchymal pattern in the images. We describe the examination of each performance by these pre-processing techniques. We show four results using (1) only magnitudes of FFT pre-processing, (2) both magnitudes and phase angles, (3) data normalized by the maximum value in the dataset, and (4) data normalized by variance of the dataset. Among the 4 pre-processing data treatments studied, the process combining FFT phase angles and magnitudes of FFT is found to be the most efficient.