Suppr超能文献

验证蒙特卡罗模拟对正常数据的三分类理想观察者工作点估计的有效性。

Validation of Monte Carlo estimates of three-class ideal observer operating points for normal data.

机构信息

The Department of Physiology, 303 E Superior St., Northwestern University, Chicago, IL 60611, USA.

出版信息

Acad Radiol. 2013 Jul;20(7):908-14. doi: 10.1016/j.acra.2013.04.002.

DOI:10.1016/j.acra.2013.04.002
PMID:23747155
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC3697838/
Abstract

RATIONALE AND OBJECTIVES

Traditional two-class receiver operating characteristic (ROC) analysis is inadequate for the complete evaluation of observer performance in tasks with more than two classes.

MATERIALS AND METHODS

Here, a Monte Carlo estimation method for operating point coordinates on a three-class ROC surface is developed and compared with analytically calculated coordinates in two special cases: (1) univariate and (2) restricted bivariate trinormal underlying data.

RESULTS

In both cases, the statistical estimates were found to be good in the sense that the analytical values lay within the 95% confidence interval of the estimated values about 95% of the time.

CONCLUSIONS

The statistical estimation method should be key in the development of a pragmatic performance metric for evaluation of observers in classification tasks with three or more classes.

摘要

原理和目的

传统的两类接收器操作特征（ROC）分析对于评估具有两个以上类别的任务中的观察者性能是不够的。

材料和方法

本文开发了一种用于三维 ROC 表面上操作点坐标的蒙特卡罗估计方法，并在两个特殊情况下与分析计算的坐标进行了比较：（1）单变量和（2）受限双变量三正态基础数据。

结果

在这两种情况下，统计估计都被发现是良好的，因为分析值大约有 95%的时间落在估计值的 95%置信区间内。

结论

对于具有三个或更多类别的分类任务中观察者性能的评估，统计估计方法应该是一种实用的性能指标的关键。

相似文献

1
Validation of Monte Carlo estimates of three-class ideal observer operating points for normal data.验证蒙特卡罗模拟对正常数据的三分类理想观察者工作点估计的有效性。
Acad Radiol. 2013 Jul;20(7):908-14. doi: 10.1016/j.acra.2013.04.002.
2
Performance analysis of three-class classifiers: properties of a 3-D ROC surface and the normalized volume under the surface for the ideal observer.三类分类器的性能分析：三维ROC曲面的特性以及理想观察者情况下该曲面下的归一化体积。
IEEE Trans Med Imaging. 2008 Feb;27(2):215-27. doi: 10.1109/TMI.2007.905822.
3
The Youden Index in the Generalized Receiver Operating Characteristic Curve Context.广义接受者操作特征曲线背景下的约登指数
Int J Biostat. 2019 Apr 3;15(1):/j/ijb.2019.15.issue-1/ijb-2018-0060/ijb-2018-0060.xml. doi: 10.1515/ijb-2018-0060.
4
Receiver Operating Characteristic (ROC) Analysis of Image Search-and-Localize Tasks.图像搜索与定位任务的受试者工作特征（ROC）分析。
Acad Radiol. 2020 Dec;27(12):1742-1750. doi: 10.1016/j.acra.2019.12.020. Epub 2020 Feb 5.
5
Optimization of restricted ROC surfaces in three-class classification tasks.三类分类任务中受限ROC曲面的优化
IEEE Trans Med Imaging. 2007 Oct;26(10):1345-56. doi: 10.1109/TMI.2007.898578.
6
Estimation receiver operating characteristic curve and ideal observers for combined detection/estimation tasks.联合检测/估计任务的估计接收器操作特性曲线和理想观察者。
J Opt Soc Am A Opt Image Sci Vis. 2007 Dec;24(12):B91-8. doi: 10.1364/josaa.24.000b91.
7
The hypervolume under the ROC hypersurface of "near-guessing" and "near-perfect" observers in N-class classification tasks.N类分类任务中“近似猜测”和“近乎完美”观察者的ROC超曲面下的超体积。
IEEE Trans Med Imaging. 2005 Mar;24(3):293-9. doi: 10.1109/tmi.2004.841227.
8
Channelized-ideal observer using Laguerre-Gauss channels in detection tasks involving non-Gaussian distributed lumpy backgrounds and a Gaussian signal.在涉及非高斯分布的块状背景和高斯信号的检测任务中，使用拉盖尔 - 高斯信道的信道化理想观察者。
J Opt Soc Am A Opt Image Sci Vis. 2007 Dec;24(12):B136-50. doi: 10.1364/josaa.24.00b136.
9
Monte Carlo validation of the Dorfman-Berbaum-Metz method using normalized pseudovalues and less data-based model simplification.使用归一化伪值和基于较少数据的模型简化对多夫曼-贝鲍姆-梅茨方法进行蒙特卡洛验证。
Acad Radiol. 2005 Dec;12(12):1534-41. doi: 10.1016/j.acra.2005.07.012.
10
Toward realistic and practical ideal observer (IO) estimation for the optimization of medical imaging systems.迈向用于医学成像系统优化的现实且实用的理想观察者（IO）估计。
IEEE Trans Med Imaging. 2008 Oct;27(10):1535-43. doi: 10.1109/TMI.2008.924641.

本文引用的文献

1
The three-class ideal observer for univariate normal data: Decision variable and ROC surface properties.单变量正态数据的三类理想观察者：决策变量和ROC曲面特性。
J Math Psychol. 2012 Aug 1;56(4):256-273. doi: 10.1016/j.jmp.2012.05.003. Epub 2012 Jun 20.
2
Nonparametric Bayesian estimation of the three-way receiver operating characteristic surface.三路接收者操作特征曲面的非参数贝叶斯估计。
Biom J. 2011 Nov;53(6):1011-24. doi: 10.1002/bimj.201100070.
3
Three-class ROC analysis--toward a general decision theoretic solution.三分类 ROC 分析——走向通用决策理论解决方案。
IEEE Trans Med Imaging. 2010 Jan;29(1):206-15. doi: 10.1109/TMI.2009.2034516. Epub 2009 Oct 30.
4
Application of three-class ROC analysis to task-based image quality assessment of simultaneous dual-isotope myocardial perfusion SPECT (MPS).三类ROC分析在同时双同位素心肌灌注单光子发射计算机断层显像（MPS）基于任务的图像质量评估中的应用。
IEEE Trans Med Imaging. 2008 Nov;27(11):1556-67. doi: 10.1109/TMI.2008.928921.
5
Performance analysis of three-class classifiers: properties of a 3-D ROC surface and the normalized volume under the surface for the ideal observer.三类分类器的性能分析：三维ROC曲面的特性以及理想观察者情况下该曲面下的归一化体积。
IEEE Trans Med Imaging. 2008 Feb;27(2):215-27. doi: 10.1109/TMI.2007.905822.
6
ROC analysis with multiple classes and multiple tests: methodology and its application in microarray studies.多类别和多测试的ROC分析：方法及其在微阵列研究中的应用
Biostatistics. 2008 Jul;9(3):566-76. doi: 10.1093/biostatistics/kxm050. Epub 2008 Feb 27.
7
Optimization of restricted ROC surfaces in three-class classification tasks.三类分类任务中受限ROC曲面的优化
IEEE Trans Med Imaging. 2007 Oct;26(10):1345-56. doi: 10.1109/TMI.2007.898578.
8
Three-class ROC analysis--the equal error utility assumption and the optimality of three-class ROC surface using the ideal observer.三类ROC分析——等错误效用假设与使用理想观察者的三类ROC曲面的最优性。
IEEE Trans Med Imaging. 2006 Aug;25(8):979-86. doi: 10.1109/tmi.2006.877090.
9
Three-class ROC analysis--a decision theoretic approach under the ideal observer framework.三类ROC分析——理想观察者框架下的一种决策理论方法。
IEEE Trans Med Imaging. 2006 May;25(5):571-81. doi: 10.1109/tmi.2006.871416.
10
Restrictions on the three-class ideal observer's decision boundary lines.对三类理想观察者决策边界线的限制。
IEEE Trans Med Imaging. 2005 Dec;24(12):1566-73. doi: 10.1109/TMI.2005.859212.

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验