Prediction of Breast Cancer by Segmenting the Image from Mammography using Neural Network Classifier
P.Suganya1, Sowmya Ramanathan2, Prachi3, CHVN Anirudh4, A.Yogendra Reddy5
1Ms. P. Suganya, Assistant Professor (O.G), Department of Computer Science and Engineering, Ramapuram Campus, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
2Ms. Sowmya Ramanathan, Student, Department of Computer Science and Engineering, Ramapuram Campus, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
3Ms. Prachi, Student, Department of Computer Science and Engineering, Ramapuram Campus, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
4Mr. CHVN Anirudh, Student, Department of Computer Science and Engineering, Ramapuram Campus, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
5Mr. A. Yogendra Reddy, Student, Department of Computer Science and Engineering, Ramapuram Campus, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
Manuscript received on 24 November 2019 | Revised Manuscript received on 07 December 2019 | Manuscript Published on 14 December 2019 | PP: 25-29 | Volume-9 Issue-1S October 2019 | Retrieval Number: A10051091S19/19©BEIESP | DOI: 10.35940/ijeat.A1005.1091S19
Open Access | Editorial and Publishing 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: An automated identification system to enable early identification of breast cancer which is one of the most familiar types of cancer amidst females which is identified using a diagnostic technique called mammography. This identification ideology banks on multiple instance learning (MIL) paradigms which demonstrate an aid in therapeutical assistance. Within the projected framework, breasts area unit is first divided adaptively into regions. The GLCM options are extracted from wavelet sub bands. A classification of diagnostic examination procedure as normal or abnormal is revealed from lesions which are masses or small calcifications and the textural options. To arrive at the final results the above factors are interpreted from all parts and analysed. In the event of an anomaly found in there port, the parts that are detected by the machine driven identification will be displayed. Dual evaluation methodology is undertaken to outline this deviation. A neural network has to be trained before utilization; it is done by segmenting the lesions and feeding it to NN. The NN assigns an anomaly index to them and then the combination of local and global anomaly index takes place.
Keywords: Multiple Instance Learning (MIL), Diagnostic Procedure, Mammography, GLCM, Computer Aided Diagnosis (CAD), Gray Scale Image.
Scope of the Article: Regression and Prediction