ARMR: Adaptively Responsive Network for Medication Recommendation

2025-07-06

AI Summary

Existing approaches struggle to dynamically balance the reuse of historical medications against the introduction of novel drugs for complex disease management. Method: We propose a segmented temporal modeling framework with an adaptive response mechanism: electronic health records are partitioned into recent and distant clinical histories to enable fine-grained temporal representation, and an attention-driven dynamic gating module adaptively modulates the weights assigned to medication reuse versus novel drug introduction based on the patient’s real-time health status. Our approach jointly leverages time-series modeling and drug co-occurrence graph learning to support personalized prescription decisions grounded in individual health evolution trajectories. Contribution/Results: Evaluated on MIMIC-III and MIMIC-IV, our method achieves statistically significant improvements over state-of-the-art baselines across Recall@10, F1-score, and diversity metrics, demonstrating strong clinical applicability and generalizability.

Technology Category

• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Planning, Routing, and Scheduling: Temporal Planning
• Machine Learning: Time-Series/Data Streams

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Medication recommendation is a crucial task in healthcare, especially for patients with complex medical conditions. However, existing methods often struggle to effectively balance the reuse of historical medications with the introduction of new drugs in response to the changing patient conditions. In order to address this challenge, we propose an Adaptively Responsive network for Medication Recommendation (ARMR), a new method which incorporates 1) a piecewise temporal learning component that distinguishes between recent and distant patient history, enabling more nuanced temporal understanding, and 2) an adaptively responsive mechanism that dynamically adjusts attention to new and existing drugs based on the patient's current health state and medication history. Experiments on the MIMIC-III and MIMIC-IV datasets indicate that ARMR has better performance compared with the state-of-the-art baselines in different evaluation metrics, which contributes to more personalized and accurate medication recommendations. The source code is publicly avaiable at: https://github.com/seucoin/armr2.

Problem

• Balancing historical and new medications for complex patients
• Adapting drug recommendations to changing patient conditions
• Improving temporal understanding of patient medication history

Innovation

• Piecewise temporal learning for history distinction
• Adaptively responsive drug attention mechanism
• Personalized medication recommendation improvement