Tumor-Track: A Machine Learning-Based Predictive Framework for Enhanced Breast Cancer Diagnosis

Main Article Content

Shilpi Singh
Govind Kumar Jha
Preetish Ranjan

Abstract

Breast cancer continues to be a worldwide health challenge with 2.3 million women diagnosed and 670,000 deaths reported in 2022. This paper would like to contribute by creating a model that uses machine learning with high precision and high efficiency in predicting breast cancer. This combines state-of-the-art algorithms in feature and clinical data analysis with a trustworthy tool to aid health professionals in screening and deciding early. The scope of this work is to give the possibility to increase the accuracy in detecting cancer, reduce false positive and offers a scalable solution applicable to mass use for breast cancer detection. The train the system to use Logistic Regression (LR) and K-Nearest Neighbors (KNN) classifiers to perform predictions. The performance is reported using the complete set of features along with a reduced subset of the ten most relevant features. Logistic Regression scores better than K-Nearest Neighbors even when working with the reduced set of the 10 most important features. The new method provides a simple but robust framework, which could help improve clinical diagnosis and patient prognosis.

Article Details

Section
Articles