Abstract:
Infectious diseases continue to pose major global health challenges, requiring rapid diagnosis, effective treatment, and timely clinical decision-making. Traditional healthcare systems often struggle with delays in identifying patient risk, selecting appropriate therapeutic options, and offering personalized care. With advancements in Artificial Intelligence (AI), recommendation systems have emerged as powerful tools that support clinicians by analyzing patient data, predicting disease progression, and suggesting optimized treatment pathways. AI-driven recommendation systems integrate electronic health records, laboratory results, and medical literature to enable intelligent decision support. Such systems improve diagnostic accuracy, reduce human error, and enhance patient outcomes, especially in infectious disease management.
This study proposes an AI-driven framework combining machine learning models, natural language processing, and multi-criteria recommendation algorithms. Patient medical histories, clinical symptoms, laboratory parameters, and genomic indicators were preprocessed and fed into classification and prediction models. Techniques such as Random Forest, XGBoost, and Deep Neural Networks were employed for disease detection and health status prediction. A hybrid recommendation engine was developed to generate personalized drug suggestions, doctor recommendations, and treatment plans. Model performance was evaluated using accuracy, F1-score, AUC, and mean reciprocal rank (MRR). The system was validated using anonymized healthcare datasets and simulated infectious disease scenarios.
The AI-driven system demonstrated high diagnostic accuracy and consistent performance across diverse infectious disease profiles. Health status prediction models achieved above 92% accuracy, enabling early identification of high-risk patients. Drug and treatment recommendation components improved therapy alignment by 30% compared to rule-based systems. Personalized doctor recommendations enhanced referral efficiency, and the integrated decision support engine significantly reduced diagnostic delays. Overall, the system improved clinical workflow efficiency and patient management outcomes.
AI-driven recommendation systems have strong potential to transform infectious disease care by enabling faster diagnosis, tailored treatment, and informed clinical decisions. The proposed framework supports healthcare professionals with real-time insights, enhances precision medicine, and promotes better patient outcomes. Such systems offer a scalable solution for addressing global healthcare challenges.
Biography:
Dr. J. Somasekar is Professor and Program Head of CSE (AIDD) at JAIN University, Bangalore, India; Research Fellow at INTI International University, Malaysia; and Head of AI & DS at UniDAIM, USA. He holds a Ph.D. from JNTUA, an M.Tech. from NIT Karnataka, and completed Postdoctoral research at the University of South Florida, USA. With 18+ years of experience, he has delivered 270+ talks across 12 countries, published 50+ papers, authored books, and guided PhD scholars. He received DST’s International Travel Grant, several global awards, and serves as Visiting Professor internationally. His research areas include Medical Imaging, AI, ML, Image Processing, Data Science, and Healthcare.
