Ovary Cancer Diagnosing Empowered with Machine Learning

Date

2022

Journal Title

Journal ISSN

Volume Title

Publisher

Institute of Electrical and Electronics Engineers Inc.

Abstract

A high mortality rate is associated with ovarian cancer, one of the most common types of cancers in women. Ovarian cancer refers to a group of disorders that develop in the ovaries and spread to the fallopian tubes and peritoneum. Treatment is most effective when ovarian cancer is discovered in its early stages. Machine learning has recently demonstrated that it is capable of better identifying ovarian cancer and its stages. Most modern research studies on ovarian cancer use a single classification model, leading to poor performance in diagnosis. For the detection of ovarian cancer, the highly sophisticated and efficient machine learning algorithms Support vector machine (SVM) and K-Nearest Neighbor (KNN) are employed in this study. Before diagnosing illness, the suggested approach can optimize and standardize data. Experimental results show that SVM has outperformed KNN in both training and validation performance and achieved an accuracy of 98.1% 97.16% for training and validation respectively. If used in medical diagnosis systems, the proposed model can significantly improve the accuracy of ovarian cancer detection leading to effective treatment and an increase in patient survival rates. © 2022 IEEE.

Description

This conference paper is not available at CUD collection. The version of scholarly record of this paper is published in 2022 International Conference on Business Analytics for Technology and Security (ICBATS) (2022), available online at: https://doi.org/10.1109/ICBATS54253.2022.9759010

Keywords

KNN, machine learning, ovary cancer, SVM

Citation

Taleb, N., Mehmood, S., Zubair, M., Naseer, I., Mago, B., & Nasir, M. U. (2022). Ovary cancer diagnosing empowered with machine learning. 2022 International Conference on Business Analytics for Technology and Security (ICBATS). https://doi.org/10.1109/ICBATS54253.2022.9759010

DOI

Collections