24 June 2024 in JMIR Medical Informatics
Background: Cirrhosis is a leading cause of non-cancer deaths in gastrointestinal diseases, resulting in high hospitalization and readmission rates. Early identification of high-risk patients is vital for proactive interventions and improve healthcare outcomes. However, the quality and integrity of real-world Electronic Health Records (EHRs) limited their utility in developing risk assessment tools. Objective: Despite the widely application of classical and ensemble machine learning for EHRs-based predictive tasks, the diversity of health conditions among patients and the inherent limitations of the data, such as incompleteness, sparsity and temporal dynamics, have not been fully addressed. To tackle those challenges, we explored a framework that characterizes patient subgroups and adaptively selects optimal predictive models for each patient on the fly to enable individualized decision support. Methods: The proposed framework uniquely addresses patient heterogeneity by aligning diverse subgroups with dynamically selected classifiers. First, patient subgroups are generated and characterized using rules indicating medical diagnosis patterns. Next, a meta-learning framework trains a meta-classifier for optimal dynamic model selection, which identifies suitable models for individual patient. Notably, we incorporated a tailored region of competence to refine model selection, specifically accounting for cirrhosis complications. This approach not only enhance predictive performance, but also elucidates why individualized predictions are better supported by selected classifiers trained on specific data subsets. Results: The proposed framework was evaluated for predicting 14-day and 30-day readmission in cirrhosis patients using multi-center data obtained from six hospitals. The final dataset comprised 3,307 patients with at least two admission records, along with a range of factors including demographic information, complications and laboratory test results. The proposed framework achieved an average AUC improvement of 5% and 4% compared to the best baseline models, respectively. Conclusions: By leveraging the expertise of the most competent classifiers for each patient subgroup, our approach enables interpretable training and dynamic selection of heterogeneous predictive models. This advancement not only improves prediction accuracy but also highlights its considerable potential for clinical applications, facilitating the alignment of diverse patient subgroups with tailored decision-support algorithms.