A Novel Approach for Classification of Mammograms using Longest Line Detection Algorithm and Decision Tree Classifier
A. M. Solanke1, Manjunath2, D. V. Jadhav3

1A.M. Solanke, Electrical & Electronics engineering, Jain (Deemed-to-be University), Bangalore, India.
2Dr. Manjunath, Electrical & Electronics engineering, Jain (Deemed-to-be-University), Bangalore, India.
3Dr. D.V.Jadhav, Joint Director, Technical Education Amaravati, India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 848-852 | Volume-8 Issue-8, June 2019 | Retrieval Number: G5236058719/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Cancer is a second leading cause of mortality in a world. Approximately 9. 6 million deaths are occurred due to cancer till 2018. As per the World Health Organisation, breast cancer is the major reason of mortality among women. Many lives can be saved by reliable detection and diagnosis of breast cancer in primitive stage. It is difficult to classify normal and abnormal mammograms accurately because of noise, dense breast tissues, unwanted parts such as labels and pectoral muscle. The methods based on Computer Aided Detection (CAD) can address these problems. These methods provide diagnosis assistance to radiologists and doctors for detection of cancer. Human errors can be reduced with the help of Computer Aided Detection algorithms. In this work mammogram images are preprocessed using longest line detection algorithm to remove pectoral muscle. Then texture and statistical features are extracted from preprocessed mammograms. Finally decision tree classifier is used to classify mammograms as normal and abnormal categories. The proposed methodology is applied to 322 mammograms. The performance analysis resulted into improved accuracy of 98.14%, sensitivity of 99.1% and specificity of 97.63%.
Keyword: Classification Mammograms Decision Organisation.
Scope of the Article: Classification.