Breast cancer risk assessment and diagnosis model using fuzzy support vector machine based expert system
- 20 January 2017
- journal article
- research article
- Published by Taylor & Francis in Journal of Experimental & Theoretical Artificial Intelligence
- Vol. 29 (5) , 1011-1021
Abstract
Classification of cancerous masses is a challenging task in many computerised detection systems. Cancerous masses are difficult to detect because these masses are obscured and subtle in mammograms. This paper investigates an intelligent classifier – fuzzy support vector machine (FSVM) applied to classify the tissues containing masses on mammograms for breast cancer diagnosis. The algorithm utilises texture features extracted using Laws texture energy measures and a FSVM to classify the suspicious masses. The new FSVM treats every feature as both normal and abnormal samples, but with different membership. By this way, the new FSVM have more generalisation ability to classify the masses in mammograms. The classifier analysed 219 clinical mammograms collected from breast cancer screening laboratory. The tests made on the real clinical mammograms shows that the proposed detection system has better discriminating power than the conventional support vector machine. With the best combination of FSVM and Laws texture features, the area under the Receiver operating characteristic curve reached .95, which corresponds to a sensitivity of 93.27% with a specificity of 87.17%. The results suggest that detecting masses using FSVM contribute to computer-aided detection of breast cancer and as a decision support system for radiologists.Keywords
This publication has 32 references indexed in Scilit:
- Texture-based analysis of clustered microcalcifications detected on mammogramsDigital Signal Processing, 2012
- Fuzzy SVM for content-based image retrieval - A pseudo-label support vector machine frameworkIEEE Computational Intelligence Magazine, 2006
- Approaches for automated detection and classification of masses in mammogramsPattern Recognition, 2006
- Support vector machines in sonographyClinical Imaging, 2005
- Computer-aided detection and classification of microcalcifications in mammograms: a surveyPattern Recognition, 2003
- Fuzzy support vector machinesIEEE Transactions on Neural Networks, 2002
- A CAD system for the automatic detection of clustered microcalcifications in digitized mammogram filmsIEEE Transactions on Medical Imaging, 2000
- An overview of statistical learning theoryIEEE Transactions on Neural Networks, 1999
- Digital mammography: mixed feature neural network with spectral entropy decision for detection of microcalcificationsIEEE Transactions on Medical Imaging, 1996
- Estimating the Error Rate of a Prediction Rule: Improvement on Cross-ValidationJournal of the American Statistical Association, 1983