Analyzing the Application of Novel Deep Learning Techniques to foresee and mitigate Breast Cancer in Women

  • P V V S Srinivas,Chandra Sekhar Kolli, Mohan Kumar Ch, Pragnyaban Mishra

Abstract

Breast cancer is one of the oldest and most critical types of cancer found in women. This type of cancer has been studied and researched by a significant number of researchers to cease consequences caused by it since it is the most deadly disease when compared to any other type. The only way to survive from such type of cancer is to detect it in the early stages and take precautions at the earliest only. The primary objective of this article is to survey the state of the art frameworks and to analyze how best we can apply and leverage the true potential of deep learning techniques to detect breast cancer in the early stage so that treatment can be initiated at the early stage itself. We have applied and thoroughly studied and analyzed the performance of the various deep learning techniques and some critical observations are drawn by analyzing the accuracy parameters such as F1-score, sensitivity, and specificity on Image and CSV dataset as well. We also applied the strategies SVM, AlexNet, GoogleNet on the image dataset and measured the performance with the same accuracy parameters. In our observation GoogleNet accuracy is improved by 9% when compared to SVM. When applied the Deep CoxPH method on image dataset it results in an improvement around 10% when compared to SVM. In overall observations, we can strongly state that the application of deep learning techniques bring out better accuracy and performance results in detecting the deadliest disease in the very beginning state itself.

Published
2020-05-01
How to Cite
P V V S Srinivas,Chandra Sekhar Kolli, Mohan Kumar Ch, Pragnyaban Mishra. (2020). Analyzing the Application of Novel Deep Learning Techniques to foresee and mitigate Breast Cancer in Women . International Journal of Advanced Science and Technology, 29(06), 5522 - 5532. Retrieved from https://sersc.org/journals/index.php/IJAST/article/view/19624