Breast Cancer Detection using Machine Learning
R. Chtihrakkannan1, P. Kavitha2, T. Mangayarkarasi3, R. Karthikeyan4
1R.Chtihrakkannan, Instrumentation and Control Engineering, Sri Sairam Engineering College, Chennai, Tamilnadu, India.
2Dr.P.Kavitha, , Electronics and Instrumentation Engineering, RMK Engineering College,Chennai, Tamilnadu, India.
3T.Mangayarkarasi, Instrumentation and Control Engineering, Sri Sairam Engineering College, Chennai, Tamilnadu, India.
4R.Karthikeyan, Instrumentation and Control Engineering, Sri Sairam Engineering College, Chennai, Tamilnadu, India.
Manuscript received on 26 August 2019. | Revised Manuscript received on 06 September 2019. | Manuscript published on 30 September 2019. | PP: 3123-3126 | Volume-8 Issue-11, September 2019. | Retrieval Number: K24980981119/2019©BEIESP | DOI: 10.35940/ijitee.K2498.0981119
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 the 2nd source of death in the world. The main reason for this increased death rate is the delayed detection of cancerous tissue growth in a person. Nearly 60% patients with breast cancer are diagnosed in advanced stages. The main objective of our paper is to enhance an image processing algorithm for earlier finding of breast cancer. X-ray mammogram images which have been acquired are used as input Images. [1] The pre-processing of input images are carried out by applying Gaussian Filter and Edge detection techniques to enhance image quality. Wavelet Transform is useful to identified first order features and GLCM based second order features are extracted from the Pre-processed images. The statistical parameters are then used for classification using DNN a Multilayer supervised classifier. Dataset images are created from the training phase. In testing Phase the acquired image from a patient is given as input to the classifier after completing the image processing steps such as Pre-processing and feature extraction. The output of the classifier consists of two classes, normal and abnormal respectively. [2] The entire algorithm is developed in Python language. The Processing time for testing and conformation of Positive cases is very minimum. Using deep learning neural network classifier an accuracy rate of 92% is reached.
Keywords: Breast Cancer, earlier detection, GLCM, DNN, Wavelet transform, accurate
Scope of the Article: Machine Learning