A Guide to ComBat Harmonization of Imaging Biomarkers in Multicenter Studies

Fanny Orlhac; Jakoba J. Eertink; Anne-S. golène Cottereau; Josée M. Zijlstra; Catherine Thieblemont; Michel Meignan; Ronald Boellaard; Irène Buvat

doi:https://doi.org/10.2967/jnumed.121.262464

A Guide to ComBat Harmonization of Imaging Biomarkers in Multicenter Studies

Fanny Orlhac, Jakoba J. Eertink, Anne-S. golène Cottereau, Josée M. Zijlstra, Catherine Thieblemont, Michel Meignan, Ronald Boellaard, Irène Buvat

Research output: Contribution to journal › Article › Academic › peer-review

84 Citations (Scopus)

Abstract

The impact of PET image acquisition and reconstruction parameters on SUV measurements or radiomic feature values is widely documented. This "scanner" effect is detrimental to the design and validation of predictive or prognostic models and limits the use of large multicenter cohorts. To reduce the impact of this scanner effect, the ComBat method has been proposed and is now used in various contexts. The purpose of this article is to explain and illustrate the use of ComBat based on practical examples. We also give examples in which the ComBat assumptions are not met; thus, ComBat should not be used.

Original language	English
Pages (from-to)	172-179
Number of pages	8
Journal	Journal of nuclear medicine : official publication, Society of Nuclear Medicine
Volume	63
Issue number	2
DOIs	https://doi.org/10.2967/jnumed.121.262464
Publication status	Published - 1 Feb 2022

Keywords

harmonization
multicenter studies
radiomics
texture analysis

Access to Document

https://doi.org/10.2967/jnumed.121.262464

Cite this

@article{7197840589b541efab9675d7d1e3ec22,

title = "A Guide to ComBat Harmonization of Imaging Biomarkers in Multicenter Studies",

abstract = "The impact of PET image acquisition and reconstruction parameters on SUV measurements or radiomic feature values is widely documented. This {"}scanner{"} effect is detrimental to the design and validation of predictive or prognostic models and limits the use of large multicenter cohorts. To reduce the impact of this scanner effect, the ComBat method has been proposed and is now used in various contexts. The purpose of this article is to explain and illustrate the use of ComBat based on practical examples. We also give examples in which the ComBat assumptions are not met; thus, ComBat should not be used.",

keywords = "harmonization, multicenter studies, radiomics, texture analysis",

author = "Fanny Orlhac and Eertink, {Jakoba J.} and Cottereau, {Anne-S. gol{\`e}ne} and Zijlstra, {Jos{\'e}e M.} and Catherine Thieblemont and Michel Meignan and Ronald Boellaard and Ir{\`e}ne Buvat",

note = "Funding Information: This work is partially supported by the Dutch Cancer Society (# VU 2018-11648). The HOVON-84 study was Funding Information: supported by the Dutch Cancer Society (CKTO-2006-14) and by Roche Nederland. All funding agencies had Publisher Copyright: {\textcopyright} 2022 Society of Nuclear Medicine Inc.. All rights reserved.",

year = "2022",

month = feb,

day = "1",

doi = "https://doi.org/10.2967/jnumed.121.262464",

language = "English",

volume = "63",

pages = "172--179",

journal = "Journal of nuclear medicine : official publication, Society of Nuclear Medicine",

issn = "1535-5667",

publisher = "Society of Nuclear Medicine Inc.",

number = "2",

}

TY - JOUR

T1 - A Guide to ComBat Harmonization of Imaging Biomarkers in Multicenter Studies

AU - Orlhac, Fanny

AU - Eertink, Jakoba J.

AU - Cottereau, Anne-S. golène

AU - Zijlstra, Josée M.

AU - Thieblemont, Catherine

AU - Meignan, Michel

AU - Boellaard, Ronald

AU - Buvat, Irène

N1 - Funding Information: This work is partially supported by the Dutch Cancer Society (# VU 2018-11648). The HOVON-84 study was Funding Information: supported by the Dutch Cancer Society (CKTO-2006-14) and by Roche Nederland. All funding agencies had Publisher Copyright: © 2022 Society of Nuclear Medicine Inc.. All rights reserved.

PY - 2022/2/1

Y1 - 2022/2/1

N2 - The impact of PET image acquisition and reconstruction parameters on SUV measurements or radiomic feature values is widely documented. This "scanner" effect is detrimental to the design and validation of predictive or prognostic models and limits the use of large multicenter cohorts. To reduce the impact of this scanner effect, the ComBat method has been proposed and is now used in various contexts. The purpose of this article is to explain and illustrate the use of ComBat based on practical examples. We also give examples in which the ComBat assumptions are not met; thus, ComBat should not be used.

AB - The impact of PET image acquisition and reconstruction parameters on SUV measurements or radiomic feature values is widely documented. This "scanner" effect is detrimental to the design and validation of predictive or prognostic models and limits the use of large multicenter cohorts. To reduce the impact of this scanner effect, the ComBat method has been proposed and is now used in various contexts. The purpose of this article is to explain and illustrate the use of ComBat based on practical examples. We also give examples in which the ComBat assumptions are not met; thus, ComBat should not be used.

KW - harmonization

KW - multicenter studies

KW - radiomics

KW - texture analysis

UR - http://www.scopus.com/inward/record.url?scp=85123972757&partnerID=8YFLogxK

U2 - https://doi.org/10.2967/jnumed.121.262464

DO - https://doi.org/10.2967/jnumed.121.262464

M3 - Article

C2 - 34531263

SN - 1535-5667

VL - 63

SP - 172

EP - 179

JO - Journal of nuclear medicine : official publication, Society of Nuclear Medicine

JF - Journal of nuclear medicine : official publication, Society of Nuclear Medicine

IS - 2

ER -

A Guide to ComBat Harmonization of Imaging Biomarkers in Multicenter Studies

Abstract

Keywords

Access to Document

Other files and links

Cite this