Personalized Medication Planning via Direct Domain Modeling and LLM-Generated Heuristics

TL;DR

The paper tackles personalized medication planning (PMP) at clinically meaningful scales by replacing domain-independent heuristics with automatically generated, problem-specific heuristics produced by large language models. It programmatically models the domain and uses GBFS guided by LLM-generated heuristics, implemented in Rust, to reason over extended PK/PD dynamics (PK concentrations and PD effects) across many drugs. Across scaled benchmark domains (including up to 28 drugs), the approach dramatically increases problem coverage and reduces planning time compared with domain-independent planners, demonstrating a clear path toward clinical decision-support deployment. The work emphasizes robustness via a retry strategy for heuristic generation under fixed resource budgets and highlights the need for further validation, explainability, and safe integration before real-world use.

Abstract

Personalized medication planning involves selecting medications and determining a dosing schedule to achieve medical goals specific to each individual patient. Previous work successfully demonstrated that automated planners, using general domain-independent heuristics, are able to generate personalized treatments, when the domain and problems are modeled using a general domain description language (\pddlp). Unfortunately, this process was limited in practice to consider no more than seven medications. In clinical terms, this is a non-starter. In this paper, we explore the use of automatically-generated domain- and problem-specific heuristics to be used with general search, as a method of scaling up medication planning to levels allowing closer work with clinicians. Specifically, we specify the domain programmatically (specifying an initial state and a successor generation procedure), and use an LLM to generate a problem specific heuristic that can be used by a fixed search algorithm (GBFS). The results indicate dramatic improvements in coverage and planning time, scaling up the number of medications to at least 28, and bringing medication planning one step closer to practical applications.

Figures (2)

• Figure 1: Per-instance comparison of the runtime of opt-hmax vs. Our Algorithm on Claude Sonnet 4.5. Points below the diagonal favor our approach. We observe a pattern where each domain's instances are approximately delineated into horizontal lines depending on which #generation solved a given instance.
• Figure 2: For medical applications it is important to have low variance. As a variance analysis we generated 30 heuristics in each domain with each model, and performed a Monte-Carlo simulation of our algorithm for 1000 iterations of sampling heuristic order, taking the coverage per-domain and overall each time. The boxes represent the first and third quartiles, the thick lines the median, and the whiskers the 1st and 99th percentiles. Although not perfect, especially in Tight(4xMeds), the method is fairly consistent, we still consistently outperform existing methods significantly, both by-domain and overall. A mitigating factor is that thanks to the formulation of heuristic search, a failure only 'returns you to square one', and does not lead to a wrong treatment.