Open AccessSegmentation and Classification of Breast Cancer Tumour
Author(s) -
Gökalp Çınarer,
Bülent Gürsel Emiroğlu,
Ahmet Haşim Yurttakal
Publication year - 2018
Publication title -
international journal of emerging trends in health sciences
Language(s) - English
Resource type - Journals
ISSN - 2547-8850
DOI - 10.18844/ijeths.v2i1.3802
Subject(s) - breast cancer , medicine , segmentation , magnetic resonance imaging , cancer , computer aided diagnosis , radiology , artificial intelligence , computer science
Breast cancer is cancer that forms in the cells of the breasts. Breast cancer is the most common cancer diagnosed in women in the world. Breast cancer can occur in both men and women, but it's far more common in women. Early detection of breast cancer tumours is crucial in the treatment. In this study, we presented a computer aided diagnosis expectation maximization segmentation and co-occurrence texture features from wavelet approximation tumour image of each slice and evaluated the performance of SVM Algorithm. We tested the model on 50 patients, among them, 25 are benign and 25 malign. The 80% of the images are allocated for training and 20% of images reserved for testing. The proposed model classified 2 patients correctly with success rate of 80% in case of 5 Fold Cross-Validation
Keywords: Breast Cancer, Computer-Aided Diagnosis (CAD), Magnetic Resonance Imaging (MRI);