BREAST CANCER PREDICTION USING DEEP LEARNING AND BIG DATA ANALYTICS ON THE VIN-DR MAMMO DATASET.
Rajesh Kumar · International Research Journal of Education and Technology (IRJET). · 2024
Breast cancer remains one of the leading causes of mortality among women worldwide, making early and accurate diagnosis critical.Traditional diagnostic approaches are timeconsuming, subjective, and prone to human error.Existing automated detection methods often struggle with small datasets, limited generalization, and computational inefficiencies.With the emergence of Big Data Analytics, there is an opportunity to improve diagnostic accuracy through the processing of large-scale medical imaging data.This study explores breast cancer prediction using Big Data Analytics and Deep Learning on the VinDr-Mammo dataset, which consists of high-resolution digitized mammogram images.We applied Convolutional Neural Networks (CNNs) for classification, achieving 95% accuracy in distinguishing malignant and benign tumors.Additionally, U-Net was utilized for tumor segmentation, enhancing detection precision.The integration of Apache Spark enabled efficient big data processing, facilitating real-time analysis and decision-making.The results demonstrate the potential of VinDr-Mammo and Big Data Analytics in automating breast cancer diagnosis, offering a reliable, scalable, and clinically applicable framework for early detection.