AI in Healthcare: Breast Cancer Prediction Using Machine Learning

Abstract

Breast cancer is still one of the top causes of death among women worldwide. This highlights the urgent need for early and reliable detection methods. Traditional diagnostic methods are often effective but can be limited by subjectivity, time demands, and the possibility of human error.    In this research, we present a machine learning-based approach for predicting breast cancer using the Breast Cancer Wisconsin dataset. The study uses a Random Forest Classifier to analyze 30 clinical features from digitized tumor images and classify cases as malignant or benign. We conducted exploratory data analysis, including class distribution and correlation heatmaps, to understand how features relate to one another.   The proposed model reached an accuracy of about 96%, with strong precision, recall, and an AUC score close to 1.0, showing its effectiveness in medical classification tasks. Feature importance analysis showed that factors like worst radius, mean concave points, and worst perimeter were key predictors.  The findings suggest that artificial intelligence can be a dependable decision-support tool for oncologists, allowing for earlier interventions and better patient outcomes. This work adds to the growing research that supports using AI in healthcare systems to improve diagnostic accuracy and efficiency.