Fuzzy‐reasoning‐based computer‐aided diagnosis for automated discrimination of myocardial heart disease from ultrasonic images

This paper describes a fuzzy‐reasoning‐based computer‐aided diagnosis scheme for automated classification of cardiomyopathy from ultrasonic images. Unlike the conventional types of membership functions such as triangle and trapezoid, Gaussian‐distributed membership functions (GDMFs) are employed in the present study. The GDMFs are initially generated using various texture‐based features computed from gray‐level co‐occurrence matrices. Subsequently, the shapes of GDMFs are optimized by a genetic‐algorithm learning process. After optimization, the classifier is used to discriminate two sets of echocardiographic images, namely, normal and abnormal cases, which were diagnosed in advance by a highly trained physician. We experimentally evaluate the performance of the proposed method against various methods reported in terms of sensitivity, specificity, and total accuracy. Our results show the promise of the proposed approach in discriminating cardiomyopathy. © 2002 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 85(11): 1–8, 2002; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/ecjc.10025