20 March 2025 in Digital Health
Background: Pediatric populations are more vulnerable to drug-induced liver injury (DILI) due to distinct pharmacokinetic profiles and ongoing physiological maturation processes. However, early identification and assessment of DILI in pediatric patients present significant clinical challenges, primarily due to the inherent complexity of pediatric cases and substantial limitations in available clinical data. Objective: This study introduces a framework that integrates clustering analysis with dynamic classifier selection (DCS) techniques to enhance pediatric DILI prediction. The proposed method addresses challenges such as patient heterogeneity and class imbalance, while optimizing predictive performance to support clinical decision-making. Methods: We investigated a retrospective cohort of 12,555 pediatric inpatients across six hospitals in Chongqing, China. The dataset encompassed a wide range of biomedical parameters, including laboratory results and liver function profiles, along with clinical documentation spanning demographic characteristics, medical histories, and medication regimens. Patients were stratified into four distinct clinical subgroups based on silhouette coefficient. A diverse pool of base classifiers was generated with varied initialization strategies and hyperparameter optimizations tailored to each patient cluster. The classification process was further refined through the implementation of Dynamic Classifier Selection with Multiple Classifier Behavior (DCS-MCB) methodology, which adaptively customizes model selection based on the distinctive clinical profiles of each subgroup. Results: The Clustering-enhanced DCS-MCB framework demonstrated superior performance compared to conventional machine learning models across evaluation metrics. The ensemble learning models consistently outperformed individual classifier models, with the presented study achieving the highest F1-score (0.926), MCC (0.917), G-mean (0.959), demonstrating the strength of this hybrid approach in addressing the complexities of pediatric DILI prediction. Conclusion: The integration of clustering analysis with dynamic classifier selection has demonstrated efficacy in complex real-world clinical settings. This methodology provides a more robust, precise, and clinically adaptable framework for patient stratification and drug safety surveillance.
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.
11 November 2026 in Biomedical Signal Processing and Control