使用机器学习来评估与北美药剂师执照考试绩效相关的因素
Douglas R Oyler1, Esther P Black1, Hope H Brandon1
1University of Kentucky, College of Pharmacy, Department of Pharmacy Practice and Science, Lexington, KY, USA.
机器学习模型准确地预测药房毕业生首次北美药剂师执照考试 (NAPLEX) 的成功. 关键预测因素包括大学考试的表现,准备软件的使用和学术历史,有助于早期识别有风险的学生.
科学领域:
- 药房教育 在药房教育.
- 药剂师执照考试 药剂师执照考试
- 机器学习在医疗保健中的应用
背景情况:
- 北美药剂师执照考试 (NAPLEX) 的药学毕业生成绩下降是一个令人担忧的问题.
- 识别NAPLEX成功的风险学生需要改进的方法.
- 机器学习 (ML) 提供了提高预测准确性的潜力.
研究的目的:
- 评估ML算法在预测首次NAPLEX通过/失败结果方面的有效性.
- 确定影响NAPLEX成功的关键学生因素.
- 将ML模型的性能与传统的物流回归进行比较.
主要方法:
- 利用了2024年肯塔基大学药学院毕业生 (n=123) 的数据.
- 评估了20多个学生特征,包括人口统计,学术历史和准备性软件参与.
- 通过CLASSify平台使用8ML算法,使用AUC-ROC用于准确性和SHAP值用于特征重要性.
主要成果:
- 四个ML算法超过了后勤回归 (AUC-ROC=0.860).
- 随机森林模型获得了最高的准确性 (AUC-ROC=0.930).
- 最好的预测功能包括大学进步考试成绩,RxPrep参与度和学业绩效指标.
结论:
- ML算法在分类NAPLEX首次性能方面表现出高准确度.
- 这些模型可以显著增强目前的战略,以识别需要支持的学生.
- 这些发现支持将ML纳入药学教育以进行积极的学生干预.
更多相关视频
09:21Author Spotlight: Generating Neuronal Phenotypic Profiles - A Protocol to Culture and Image Human Midbrain Dopaminergic Neurons
Published on: July 7, 2023
07:31A Computerized Functional Skills Assessment and Training Program Targeting Technology Based Everyday Functional Skills
Published on: February 13, 2020
相关概念视频
Analysis of Population Pharmacokinetic Data
Factors Affecting Drug Distribution: Miscellaneous Factors
Age plays a significant role due to differences in body composition among different age groups. Infants, for instance, have a higher proportion of total body water and lower albumin levels, a protein that binds drugs in the bloodstream. This unique composition in infants enhances the...
Factors Affecting Drug Response: Overview
Factors Affecting Protein-Drug Binding: Patient-Related Factors
Age stands as a key determinant in protein-drug binding. Neonates, characterized by low albumin content, experience heightened concentrations of unbound drugs such as phenytoin and...
Factors Affecting Protein-Drug Binding: Drug-Related Factors
One crucial factor in drug-protein binding is the drug's lipophilicity or its affinity for fat. More lipophilic drugs tend to have higher binding extents. For example, highly lipophilic drugs like cloxacillin exhibit substantial protein binding, with as much as 95% of the drug binding to proteins. In...
Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
