Deep Learning Model for Detection of Breast Cancer

7 Pages Posted: 7 May 2019

See all articles by Varun Sapra

Varun Sapra

Jagannath University

Madan Lal Saini

Jagannath University

Date Written: March 14, 2019

Abstract

Breast cancer is a type of cancer that has highest mortality rate in women in almost all the nations, accounting for 14% of all cancer deaths worldwide. Early and accurate diagnosis can help medical practitioners to take early decisions that can protect a large number of patients from unnecessary medical expenses and can provide them with early treatment. There are some widely accepted techniques or methods that are being used by the clinical experts for the diagnosing of breast cancer. Some of them are biopsy, mammography and Fine Needle Aspirate. The emergence of machine learning algorithms and accessibility to health care data offers more opportunities for comprehensive analysis and identification of diseases non-invasively much accurately and hence can improve treatment and prevention of diseases like breast cancer. The proposed study provides another novel method for the identification of breast cancer using a multilayer perceptron feed forward network. For experiment purpose, we used the breast cancer related data available at UC Irvine machine-learning repository. The model achieved the accuracy of 98.43%.

Keywords: Classification, Feed Forward Network, Prediction System, Non Invasive detection

Suggested Citation

Sapra, Varun and Lal Saini, Madan, Deep Learning Model for Detection of Breast Cancer (March 14, 2019). International Conference on Advances in Engineering Science Management & Technology (ICAESMT) - 2019, Uttaranchal University, Dehradun, India, Available at SSRN: https://ssrn.com/abstract=3383336 or http://dx.doi.org/10.2139/ssrn.3383336

Varun Sapra (Contact Author)

Jagannath University ( email )

India

Madan Lal Saini

Jagannath University ( email )

India

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
24
Abstract Views
231
PlumX Metrics