An Intelligent Real-Time Breast Cancer Classification in a Cloud Environment Using Optimized ELM

Abstract

Breast cancer remains a major global health challenge, particularly for women in remote regions where the medical infrastructure is very poor. AI-powered diagnostic systems supported by cloud computing technology offer a practical way to deliver early detection services to these under-served areas. This work proposes an AI-driven framework that combines Gain Ratio–based feature selection, PCA for reducing overfitting, an optimized Extreme Learning Machine (ELM) for breast cancer classification, and a cloud-enabled platform for remote diagnosis. Using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset, the cloud-based ELM is compared with leading diagnostic methods. Results show that the proposed system outperforms existing approaches, achieving 98.25 % accuracy, 0.9891 recall, 0.9861 precision, an F1-score of 0.9861, and an AUC of 0.9931.