阿片类药物使用障碍预测中用于电子健康记录诊断代码的特征选择方法对比研究
A Comparative Study of Feature Selection Methods for EHR Diagnosis Codes in Opioid Use Disorder Prediction
浏览论文内容
中文总结 AI 辅助
本研究对比五种特征选择方法在阿片类药物使用障碍预测中的表现，发现NTK敏感性方法在准确性与稳定性间平衡最优，LLM引导选择可提供互补临床信号。
中文摘要 AI 辅助
特征选择是基于电子健康记录（EHR）的预测建模中的关键步骤，这类建模的输入变量通常具有高维、稀疏、噪声大且冗余的特点。大型特征集不仅会增加计算负担和过拟合风险，还会使模型解释变得困难，导致其在临床环境中的实用性受限。本研究聚焦于与诊断相关的特征，对比五种用于阿片类药物使用障碍（OUD）预测的特征选择范式：复发富集、基于神经切线核（NTK）的早期梯度敏感性、LightGBM-SHAP、弹性网（Elastic Net）以及大语言模型（LLM）引导的语义选择。我们采用统一的预处理与评估框架，通过下游预测性能、重采样稳定性以及对罕见诊断代码的表征来评估每种方法。结果表明，随着特征预算增大，性能会提升，但在达到中等规模后会出现收益递减；NTK敏感性在准确性与稳定性之间提供了最佳平衡，而LLM引导的选择尽管独立性能较低，仍能提供互补的、具有临床意义的信号。
英文摘要
Feature selection is a critical step in electronic health record (EHR)-based predictive modeling, where input variables are often high-dimensional, sparse, noisy, and redundant. Large feature sets not only increase computational burden and overfitting risk, but also make model interpretation difficult, leading to limited usefulness in clinical settings. In this study, we focus on diagnosis-related features and compare five feature selection paradigms for opioid use disorder (OUD) prediction: recurrence enrichment, NTK-motivated early gradient sensitivity, LightGBM-SHAP, Elastic Net, and large language model (LLM)-guided semantic selection. We use a unified preprocessing and evaluation framework and assess each method by downstream predictive performance, resampling stability, and representation of infrequent diagnosis codes. Our results demonstrate that performance improves with larger feature budgets with diminishing returns beyond a moderate size. NTK sensitivity provides the best overall balance of accuracy and stability, and LLM-guided selection contributes complementary clinically meaningful signals despite lower standalone performance.
发表机构
- Stony Brook University（石溪大学）
- Rutgers University（罗格斯大学）
- University of Vermont（佛蒙特大学）
机构由 AI 辅助整理，请以论文原文为准。
