To search, Click below search items.


All Published Papers Search Service


Comparison Machine Learning Algorithms in Abnormal Mammograms Classification


Youssef Ben Youssef, Elhassane Abdelmounim, Jamal Zbitou, Mustapha Elharoussi, Mohammed Najib Boujida


Vol. 17  No. 5  pp. 19-25


Breast cancer is the first women health problem in the word. Early detection of breast cancer is the golden key for reduction of mortality. All the radiologists are in front of an interpretation and a decision making on a mammographic image. It has been shown that in current breast cancer screenings 8%?20% of the tumors are missed by the radiologists. Computer aided diagnosis (CAD) system in mammography diagnosis can be used to interpret mammography image and make decision. Thus classification into malignant and benign tumours of breast cancer is done using machine learning techniques. In this paper, we present an efficient computer aided mammogram classification using Machine Learning Techniques (MLT) like Multi-Layer Perceptron (MLP) and Support Vector Machine (SVM). In this work suspicious mass recognition is its shape and the last is used as a powerful descriptor to distinguish between malignant and benign mammogram. ChanVese model for segmentation is used due to its robustness and non sensitive to noise. Once Region Of Interest (ROI) is segmented then local descriptors (shape) are extracted and use it in automatique classification. In order to validate our proposed method, the Moroccan mammographic image databases are used. Two classifiers SVM and MLP, derived on machine learning techniques, are used and compared their performances in term of accuracy. Results show that the SVM classifier based on local descriptors(shape) gives a satisfactory accuracy compared to MLP.


Computer Aided Diagnosis (CAD), Machine Learning Techniques, Support Vector Machine, Multi-Layer Perceptron, Segmentation