AI Technologies for Predicting Susceptibility and Outcomes of Immunological Disorders
R Keerthana, Srinivasan Savitha, K. Logeswaran, A. Rajivkannan, M. Namasivayam · Advances in computational intelligence and robotics book series · 2025
AI improves immune disease prediction by examining clinical and genomic information to determine biomarkers, enhance diagnoses, individualize treatment, and predict outcomes. Challenges include scarce high-quality data, integrating genomic and environmental variables, privacy legislation, bias, and deep learning transparency lowering clinician confidence. Federated learning maintains privacy but grapples with data quality, and explainable AI (XAI) assists interpretability but remains clinically irrelevant. SVMs, RF, CNNs, and RNNs are AI models that help in disease tracking and diagnosis. Clustering and Bayesian networks evaluate risks. Maximizing the potential of AI requires enhancing data quality, model interpretability, and clinician-data scientist collaboration to improve immunological disorder prediction.