Title :
A PC-based tissue classification system using artificial neural networks
Author :
Botros, Nazeih M.
Author_Institution :
Dept. of Electr. Eng., Southern Illinois Univ., Carbondale, IL, USA
fDate :
10/1/1992 12:00:00 AM
Abstract :
The author presents a pattern recognition algorithm and describes required instrumentation for ultrasound classification of simulated human-liver tissue abnormalities. The tissue is simulated by a liver phantom that mimics the tissue acoustically. The instrumentation used, a 50-MHz microcomputer-based data acquisition and analysis system designed by the author, digitizes the ultrasound backscattered signal from selected regions of the phantom and processes the digitized data for feature measurement. The algorithm is based on a three-layer backpropagation artificial neural network; trained to differentiate between simulated normal and abnormal tissue and to classify three types of simulated abnormalities. The results show that out of 28 cases the system classifies 25 correctly and fails to classify three cases. The reasons for this are discussed along with recommendations to increase the accuracy of classification
Keywords :
backpropagation; backscatter; biomedical equipment; biomedical ultrasonics; data acquisition; digital simulation; feature extraction; image recognition; liver; medical image processing; microcomputer applications; neural nets; ultrasonic scattering; 50 MHz; artificial neural networks; data acquisition; digitized data; feature measurement; liver phantom; microcomputer; pattern recognition algorithm; simulated human-liver tissue abnormalities; three-layer backpropagation artificial neural network; tissue classification; ultrasound backscattered signal; ultrasound classification; Data acquisition; Data analysis; Imaging phantoms; Instruments; Liver; Pattern recognition; Signal analysis; Signal design; Signal processing; Ultrasonic imaging;
Journal_Title :
Instrumentation and Measurement, IEEE Transactions on
