Multimodal AI for Hospital Readmission Prediction Among Older Adults

Predicting hospital readmissions among elderly patients is a significant concern for healthcare systems due to its impact on patient well-being, financial costs, and clinical workload. This paper investigates the role of multimodal artificial intelligence (AI) in forecasting readmission risks by integrating diverse information sources such as electronic health records (EHRs), diagnostic images, clinical notes, and demographic attributes. Unlike traditional single-source models, multimodal AI leverages both structured and unstructured data to uncover complex relationships influencing patient outcomes. The study reviews key architectures including fusion-based deep learning networks, recurrent neural networks (RNNs), long short-term memory networks (LSTMs), and transformer-based frameworks that effectively process temporal and contextual patient data. Data preprocessing techniques such as normalization, feature selection, and missing-value handling are discussed to ensure reliable predictions. Evaluation metrics including accuracy, AUC, precision, recall, and F1-score are applied to assess performance. Major challenges include data heterogeneity, privacy concerns, and limited model interpretability. Future research directions focus on explainable AI, privacy-preserving learning, and large-scale multimodal healthcare datasets. Overall, multimodal AI demonstrates strong potential to improve hospital readmission prediction and support preventive healthcare strategies for older adults.