Automated Breast Cancer Detection Using Machine Learning Techniques by Extracting Different Feature Extracting Strategies

• Published in: 2018 17th IEEE International Conference On Trust, Security And Privacy In Computing And Communications/ 12th IEEE International Conference On Big Data Science And Engineering (TrustCom/BigDataSE) pp. 327 - 331
• Main Authors: Hussain, Lal, Aziz, Wajid, Saeed, Sharjil, Rathore, Saima, Rafique, Muhammad
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2018
• Subjects: Breast cancer; Classification; Decision trees; Entropy; Feature extraction; Feature extration; Machine Learning; Mammography; Support vector machines
• ISSN: 2324-9013
• DOI: 10.1109/TrustCom/BigDataSE.2018.00057

This Breast Cancer in women is the most frequency diagnosed and second leading cause of cancer deaths. Due to complex nature of microcalcification and masses, radiologist fail to properly diagnose breast cancer. In past researchers developed Computer aided diagnosis (CAD) systems that help the radiologist to detect abnormalities in an efficient manner. In this research, we have employed robust Machine learning classification techniques such as Support vector machine (SVM) kernels and Decision Tree to distinguish cancer mammograms from normal subjects. Different features are proposed such as texture, morphological entropy based, scale invariant feature transform (SIFT), and elliptic Fourier descriptors (EFDs). These features are passed as input to ML classifiers. Jack-knife 10-fold cross validation was used and performance evaluated in term of specificity, sensitivity, Positive predive value (PPV), negative predictive value (NPV), false positive rate (FPR) and receive operating curve (ROC). The highest performance based on single feature extracting strategy was obtained using Bayesian approach with texture and EFDs features, and SVM RBF and Gaussian kernels with EFDs features whereas highest AUC with single feature was obtained using Bayesian approach by extracting texture, morphological, EFDs and entropy features and SVM RBF and Gaussian kernels with EFDs features.