Performance Evaluation of Multi-Disease Classification and Recommendation Models

This paper aims to review the current advances in healthcare uses of machine learning (ML) for the diagnosis of numerous diseases using different classification models. It proposes an efficient mechanism for multi-disease diagnosis by implementing the Bee-Featured Convolutional Neural Network (BFCNN), coupled with a Hybrid Statistical Fuzzy (Hybrid S-Fuzzy) scheme, to improve classification efficiency. The primary goal is to develop an accurate diagnostic system for heart, kidney, and liver diseases at early stages. The methodology presented compares several common ML models-Naive Bayes (NB), Support Vector Machine (SVM), Decision Tree (DT), and Fast Condition Mutual Information (FCMIM-SVM)-using datasets from various sources with the BFCNN model. According to the simulated results, the proposed BFCNN model achieved a 96.7% accuracy rate and outperformed existing models in terms of sensitivity and specificity. This study suggests that using the BFCNN together with the Hybrid S-Fuzzy model enhances diagnosis for multiple diseases, offering higher accuracy and earlier identification compared to individual models, with up to 78% potential benefits in healthcare. Further studies on this model may lead to the application of deeper methods to increase diagnostic accuracy and improve treatment recommendations.