Enhanced Medical Intelligent System for Cancer Disease Prediction, Using Case-Based Reasoning

Abstract

Breast cancer remains a leading cause of morbidity and mortality among women globally, necessitating timely and accurate diagnostic support systems. This research proposes a hybrid intelligent system that integrates Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Case-Based Reasoning (CBR) for early breast cancer prediction and diagnosis. The model was trained on a clinical breast cancer dataset, with feature attributes such as tumor size, symptom severity, hormone receptor status, HER2 status, and other relevant features. To ensure high-quality input, rigorous data preprocessing and feature engineering techniques were applied, including encoding of categorical variables, feature scaling, and exploratory analysis. The hybrid SVM-KNN model demonstrated high predictive performance, achieving an accuracy of 95.3%, along with strong precision, recall, and F1-score metrics. The system was deployed using Streamlit and integrated with an SQLite database, enabling real-time predictions, case-based retrieval, and diagnostic report generation.