Modern clinical practice relies on evidence-based guidelines implemented as compact scoring systems composed of a small number of interpretable decision rules. While machine-learning models achieve strong performance, many fail to translate into routine clinical use due to misalignment with workflow constraints such as memorability, auditability, and bedside execution. We argue that this gap arises not from insufficient predictive power, but from optimizing over model classes that are incompatible with guideline deployment. Deployable guidelines often take the form of unit-weighted clinical checklists, formed by thresholding the sum of binary rules, but learning such scores requires searching an exponentially large discrete space of possible rule sets. We introduce AgentScore, which performs semantically guided optimization in this space by using LLMs to propose candidate rules and a deterministic, data-grounded verification-and-selection loop to enforce statistical validity and deployability constraints. Across eight clinical prediction tasks, AgentScore outperforms existing score-generation methods and achieves AUROC comparable to more flexible interpretable models despite operating under stronger structural constraints. On two additional externally validated tasks, AgentScore achieves higher discrimination than established guideline-based scores.

翻译：现代临床实践依赖于以少量可解释决策规则构成的紧凑评分系统形式实施的循证指南。尽管机器学习模型表现出色，但由于与临床工作流约束（如可记忆性、可审计性和床旁执行能力）不匹配，许多模型未能转化为常规临床应用。我们认为这一差距并非源于预测能力不足，而是因为其优化时所使用的模型类别与指南部署不相兼容。可部署的指南通常采用单位加权临床检查表的形式，通过对二值规则求和并设定阈值而构建，但学习此类评分需要在可能的规则集构成的指数级离散空间中搜索。我们提出AgentScore方法，该方法在此空间中利用LLM进行语义引导的优化：通过LLM提出候选规则，并采用确定性、基于数据验证与筛选的闭环机制来保证统计有效性和可部署性约束。在八项临床预测任务中，AgentScore优于现有评分生成方法，尽管在更强的结构约束下运行，其AUROC仍可与更灵活的可解释模型相媲美。在另外两项外部验证任务中，AgentScore的分辨能力优于既有的基于指南的评分标准。