Development of a Web-Based Fuzzy Expert System for Breast Cancer Risk Assessment

Abstract

Breast cancer is a leading cause of morbidity and mortality among women worldwide, underscoring the urgent need for reliable, accessible, and cost-effective risk assessment tools. This study developed a Web-Based Fuzzy Expert System (WFES) for breast cancer risk prediction, designed to integrate clinical risk factors into an intelligent decision-support platform. Eight variables—age, age at first and last menstrual cycle, age at first pregnancy, duration of breastfeeding, body mass index, alcohol intake, and smoking—were represented using fuzzy membership functions. A knowledge base of 50 IF–THEN rules, constructed from medical expert input, was implemented through a Mamdani inference engine and deployed via a web application for platform-independent accessibility. The WFES classified patients into five risk levels: very low, low, moderate, high, and very high. Evaluation with 50 patient datasets revealed strong performance, with a Cohen’s Kappa coefficient of 0.885, indicating almost perfect agreement with human expert assessments. Beyond its predictive accuracy, the web-based architecture ensures broad usability and scalability, particularly in resource-limited healthcare settings. This study demonstrates that the WFES can serve as a practical and interpretable decision-support system, aiding both clinicians and patients in early breast cancer risk assessment and prevention planning.