Comparison of ROC methods for partially-paired data

Brandon D. Gallas; Lorenzo L. Pesce

doi:10.1117/12.813688

Comparison of ROC methods for partially-paired data

Brandon D. Gallas, Lorenzo L. Pesce

Pharmacology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

In this work we investigate ROC methods that compare the difference in AUCs (area under the ROC curve) from two modalities given partially paired data. Such methods are needed to accommodate the real world situations, where every case cannot be imaged or interpreted using both modalities. We compare variance estimation of the bivariate binormal-model based method ROCKIT of Metz et al., as well as several different non-parametric methods, including the bootstrap and U-statistics. This comparison explores different ROC curves, study designs (pairing structure of the data), sample sizes, case mix, and modality effect sizes.

Original language	English (US)
Title of host publication	Medical Imaging 2009
Subtitle of host publication	Image Perception, Observer Performance, and Technology Assessment
DOIs	https://doi.org/10.1117/12.813688
State	Published - 2009
Event	Medical Imaging 2009: Image Perception, Observer Performance, and Technology Assessment - Lake Buena Vista, FL, United States Duration: Feb 11 2009 → Feb 12 2009

Publication series

Name	Progress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume	7263
ISSN (Print)	1605-7422

Conference

Conference	Medical Imaging 2009: Image Perception, Observer Performance, and Technology Assessment
Country/Territory	United States
City	Lake Buena Vista, FL
Period	2/11/09 → 2/12/09

Keywords

AUC
Bootstrap
Missing data
Partially-paired data
ROCKIT
Receiver operating characteristic (ROC)

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
Atomic and Molecular Physics, and Optics
Biomaterials
Radiology Nuclear Medicine and imaging

Access to Document

10.1117/12.813688

Cite this

@inproceedings{042c22804fda45978f781131bb6d52bc,

title = "Comparison of ROC methods for partially-paired data",

abstract = "In this work we investigate ROC methods that compare the difference in AUCs (area under the ROC curve) from two modalities given partially paired data. Such methods are needed to accommodate the real world situations, where every case cannot be imaged or interpreted using both modalities. We compare variance estimation of the bivariate binormal-model based method ROCKIT of Metz et al., as well as several different non-parametric methods, including the bootstrap and U-statistics. This comparison explores different ROC curves, study designs (pairing structure of the data), sample sizes, case mix, and modality effect sizes.",

keywords = "AUC, Bootstrap, Missing data, Partially-paired data, ROCKIT, Receiver operating characteristic (ROC)",

author = "Gallas, {Brandon D.} and Pesce, {Lorenzo L.}",

year = "2009",

doi = "10.1117/12.813688",

language = "English (US)",

isbn = "9780819475145",

series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

booktitle = "Medical Imaging 2009",

note = "Medical Imaging 2009: Image Perception, Observer Performance, and Technology Assessment ; Conference date: 11-02-2009 Through 12-02-2009",

}

Gallas, BD & Pesce, LL 2009, Comparison of ROC methods for partially-paired data. in Medical Imaging 2009: Image Perception, Observer Performance, and Technology Assessment., 72630V, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 7263, Medical Imaging 2009: Image Perception, Observer Performance, and Technology Assessment, Lake Buena Vista, FL, United States, 2/11/09. https://doi.org/10.1117/12.813688

TY - GEN

T1 - Comparison of ROC methods for partially-paired data

AU - Gallas, Brandon D.

AU - Pesce, Lorenzo L.

PY - 2009

Y1 - 2009

N2 - In this work we investigate ROC methods that compare the difference in AUCs (area under the ROC curve) from two modalities given partially paired data. Such methods are needed to accommodate the real world situations, where every case cannot be imaged or interpreted using both modalities. We compare variance estimation of the bivariate binormal-model based method ROCKIT of Metz et al., as well as several different non-parametric methods, including the bootstrap and U-statistics. This comparison explores different ROC curves, study designs (pairing structure of the data), sample sizes, case mix, and modality effect sizes.

AB - In this work we investigate ROC methods that compare the difference in AUCs (area under the ROC curve) from two modalities given partially paired data. Such methods are needed to accommodate the real world situations, where every case cannot be imaged or interpreted using both modalities. We compare variance estimation of the bivariate binormal-model based method ROCKIT of Metz et al., as well as several different non-parametric methods, including the bootstrap and U-statistics. This comparison explores different ROC curves, study designs (pairing structure of the data), sample sizes, case mix, and modality effect sizes.

KW - AUC

KW - Bootstrap

KW - Missing data

KW - Partially-paired data

KW - ROCKIT

KW - Receiver operating characteristic (ROC)

UR - http://www.scopus.com/inward/record.url?scp=67249134163&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67249134163&partnerID=8YFLogxK

U2 - 10.1117/12.813688

DO - 10.1117/12.813688

M3 - Conference contribution

AN - SCOPUS:67249134163

SN - 9780819475145

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2009

T2 - Medical Imaging 2009: Image Perception, Observer Performance, and Technology Assessment

Y2 - 11 February 2009 through 12 February 2009

ER -

Comparison of ROC methods for partially-paired data

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this