相关实验视频

Updated: Jul 7, 2025

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

半监督的ROC分析可用于可靠和精简的表型算法评估

Jianhui Gao¹, Clara-Lea Bonzel², Chuan Hong³

¹Department of Statistical Sciences, University of Toronto, Toronto, ON, Canada.

Journal of the American Medical Informatics Association : JAMIA

|December 21, 2023

概括

一种新的半监督方法 (ssROC) 精确地使用最小标记的电子健康记录 (EHR) 数据评估表型算法 (PA). 与传统的监督方法相比,这种方法显著减少了绩效估计的差异,简化了翻译研究.

科学领域:

生物医学信息学生物医学信息学
医疗保健中的机器学习
翻译研究是翻译研究.

背景情况:

使用电子健康记录 (EHR) 的高通量表型化对于翻译研究至关重要.
估计和评估表型算法 (PA) 需要广泛的医疗监督,这构成了一个重要的瓶.
现有的弱监督的学习方法缺乏可靠的评估策略,对于具有有限标记数据的PA.

研究的目的:

引入一种新的半监督方法ssROC,用于估计PAs的接收器运行特征 (ROC) 参数.
以很小比例的标记数据来解决评估PA预测性能的挑战.
为了能够精确估计PA性能指标,如灵敏度和特异性.

主要方法:

ssROC使用一个小的标记数据集来非参数计算缺失的标签.
然后用于ROC参数估计的imputed标签.
使用合成,半合成和现实世界EHR数据评估了ssROC的性能.

主要成果:

与传统的监督ROC分析 (supROC) 相比,sssROC在ROC参数估计中显示出最小的偏差和显著较低的差异.
在模拟中,ssROC在精度上胜过supROC.
对于来自Mass General Brigham (MGB) EHR数据的5个PA,ssROC估计的平均变化比supROC少30%至60%.

结论:

ssROC可以精确评估PA的性能,而不需要大量的标记数据.
使用开源R软件,ssROC方法很容易实现.
将ssROC与弱监督的PA集成起来,为研究提供了可靠和简化的基于EHR的表型化.

关键词:

在ROC分析中,ROC分析电子健康记录是电子健康记录.现型定制现型定制现型定制在半监督状态下.

更多相关视频

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education

Published on: August 16, 2024

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

相关概念视频

Receiver Operating Characteristic Plot

Receiver Operating Characteristic Plot

A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...