Breast Cancer Detection Using Machine Learning

Breast Cancer Detection Using Machine Learning

Vasi Hussain Sayed1, Atharva Raut2, Hemlata Patil3, Niharika Ranjerla4, Prof. Chandrakant Rane5
1,2,3,4,5 COMPUTER ENGINEERING, INDALA COLLEGE OF ENGINEERING, KALYAN

---------------------------------------------------------------------***--------------------------------------------------------------------

Abstract
Breast cancer continues to be a major health challenge for women around the world, with countless new cases emerging each year. Catching the disease early makes a significant difference in treatment success and survival rates. However, traditional diagnostic tools—though helpful—can fall short due to time delays, human error, or limited precision. In this study, we looked into how machine learning (ML) techniques can support tumor classification by drawing insights from both clinical records and medical images. We experimented with algorithms like Support Vector Machines (SVM), Random Forest to access their performance in classifying tumors. Using the CBIS-DDSM: Breast Cancer Image Dataset as a training base, we developed a system that could potentially support real-time clinical decisions. The dataset was split using train_test_split to ensure unbiased evaluation. The SVM was fine-tuned using cross-validation to find the best parameters, significantly enhancing accuracy. Results indicated strong classification performance in terms of precision, recall, and F1-score. This highlights SVM’s potential in medical image analysis. Our goal wasn’t just to showcase the technology but to demonstrate how ML could serve as a practical tool in modern diagnostics, offering faster, more consistent, and scalable solutions in cancer detection.