Pediatric brain tumor consortium multisite assessment of apparent diffusion coefficient z-Axis variation assessed with an ice-water phantom

Robert V. Mulkern, Kelsey I. Ricci, Sridhar Vajapeyam, Thomas L. Chenevert, Dariya I. Malyarenko, Mehmet Kocak, Tina Young Poussaint

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Rationale and Objectives: Magnetic resonance diffusion imaging can characterize physiologic characteristics of pediatric brain tumors used to assess therapy response. The purpose of this study was to assess the variability of the apparent diffusion coefficient (ADC) along z-axis of scanners in the multicenter Pediatric Brain Tumor Consortium (PBTC). Materials and Methods: Ice-water diffusion phantoms for each PBTC site were distributed with a specific diffusion imaging protocol. The phantom was scanned four successive times to 1) confirm water in the tube reached thermal equilibrium and 2) allow for assessment of intra-examination ADC repeatability. ADC profiles across slice positions for each vendor and institution combination were characterized using linear regression modeling with a quadratic fit. Results: Eleven sites collected data with a high degree of compliance to the diffusion protocol for each scanner. The mean ADC value at slice position zero for vendor A was 1.123×10-3mm2/s, vendor B was 1.0964×10-3mm2/s, and vendor C was 1.110×10-3mm2/s. The percentage coefficient of variation across all sites was 0.309% (standard deviation=0.322). The ADC values conformed well to a second-order polynomial along the z-axis, (ie, following a linear model pattern with quadratic fit) for vendor-institution combinations and across vendor-institution combinations as shown in the longitudinal model. Conclusions: Assessment of the variability of diffusion metrics is essential for establishing the validity of using these quantitative metrics in multicenter trials. The low variability in ADC values across vendors and institutions and validates the use of ADC as a quantitative tumor marker in pediatric multicenter trials.

Original languageEnglish (US)
Pages (from-to)363-369
Number of pages7
JournalAcademic Radiology
Volume22
Issue number3
DOIs
StatePublished - Jan 1 2015

Fingerprint

Ice
Brain Neoplasms
Pediatrics
Water
Multicenter Studies
Linear Models
Diffusion Magnetic Resonance Imaging
Tumor Biomarkers
Hot Temperature

All Science Journal Classification (ASJC) codes

  • Radiology Nuclear Medicine and imaging

Cite this

Pediatric brain tumor consortium multisite assessment of apparent diffusion coefficient z-Axis variation assessed with an ice-water phantom. / Mulkern, Robert V.; Ricci, Kelsey I.; Vajapeyam, Sridhar; Chenevert, Thomas L.; Malyarenko, Dariya I.; Kocak, Mehmet; Poussaint, Tina Young.

In: Academic Radiology, Vol. 22, No. 3, 01.01.2015, p. 363-369.

Research output: Contribution to journalArticle

Mulkern, Robert V. ; Ricci, Kelsey I. ; Vajapeyam, Sridhar ; Chenevert, Thomas L. ; Malyarenko, Dariya I. ; Kocak, Mehmet ; Poussaint, Tina Young. / Pediatric brain tumor consortium multisite assessment of apparent diffusion coefficient z-Axis variation assessed with an ice-water phantom. In: Academic Radiology. 2015 ; Vol. 22, No. 3. pp. 363-369.
@article{ffc626a6c8c74401b926fd45f29f65b5,
title = "Pediatric brain tumor consortium multisite assessment of apparent diffusion coefficient z-Axis variation assessed with an ice-water phantom",
abstract = "Rationale and Objectives: Magnetic resonance diffusion imaging can characterize physiologic characteristics of pediatric brain tumors used to assess therapy response. The purpose of this study was to assess the variability of the apparent diffusion coefficient (ADC) along z-axis of scanners in the multicenter Pediatric Brain Tumor Consortium (PBTC). Materials and Methods: Ice-water diffusion phantoms for each PBTC site were distributed with a specific diffusion imaging protocol. The phantom was scanned four successive times to 1) confirm water in the tube reached thermal equilibrium and 2) allow for assessment of intra-examination ADC repeatability. ADC profiles across slice positions for each vendor and institution combination were characterized using linear regression modeling with a quadratic fit. Results: Eleven sites collected data with a high degree of compliance to the diffusion protocol for each scanner. The mean ADC value at slice position zero for vendor A was 1.123×10-3mm2/s, vendor B was 1.0964×10-3mm2/s, and vendor C was 1.110×10-3mm2/s. The percentage coefficient of variation across all sites was 0.309{\%} (standard deviation=0.322). The ADC values conformed well to a second-order polynomial along the z-axis, (ie, following a linear model pattern with quadratic fit) for vendor-institution combinations and across vendor-institution combinations as shown in the longitudinal model. Conclusions: Assessment of the variability of diffusion metrics is essential for establishing the validity of using these quantitative metrics in multicenter trials. The low variability in ADC values across vendors and institutions and validates the use of ADC as a quantitative tumor marker in pediatric multicenter trials.",
author = "Mulkern, {Robert V.} and Ricci, {Kelsey I.} and Sridhar Vajapeyam and Chenevert, {Thomas L.} and Malyarenko, {Dariya I.} and Mehmet Kocak and Poussaint, {Tina Young}",
year = "2015",
month = "1",
day = "1",
doi = "10.1016/j.acra.2014.10.006",
language = "English (US)",
volume = "22",
pages = "363--369",
journal = "Academic Radiology",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "3",

}

TY - JOUR

T1 - Pediatric brain tumor consortium multisite assessment of apparent diffusion coefficient z-Axis variation assessed with an ice-water phantom

AU - Mulkern, Robert V.

AU - Ricci, Kelsey I.

AU - Vajapeyam, Sridhar

AU - Chenevert, Thomas L.

AU - Malyarenko, Dariya I.

AU - Kocak, Mehmet

AU - Poussaint, Tina Young

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Rationale and Objectives: Magnetic resonance diffusion imaging can characterize physiologic characteristics of pediatric brain tumors used to assess therapy response. The purpose of this study was to assess the variability of the apparent diffusion coefficient (ADC) along z-axis of scanners in the multicenter Pediatric Brain Tumor Consortium (PBTC). Materials and Methods: Ice-water diffusion phantoms for each PBTC site were distributed with a specific diffusion imaging protocol. The phantom was scanned four successive times to 1) confirm water in the tube reached thermal equilibrium and 2) allow for assessment of intra-examination ADC repeatability. ADC profiles across slice positions for each vendor and institution combination were characterized using linear regression modeling with a quadratic fit. Results: Eleven sites collected data with a high degree of compliance to the diffusion protocol for each scanner. The mean ADC value at slice position zero for vendor A was 1.123×10-3mm2/s, vendor B was 1.0964×10-3mm2/s, and vendor C was 1.110×10-3mm2/s. The percentage coefficient of variation across all sites was 0.309% (standard deviation=0.322). The ADC values conformed well to a second-order polynomial along the z-axis, (ie, following a linear model pattern with quadratic fit) for vendor-institution combinations and across vendor-institution combinations as shown in the longitudinal model. Conclusions: Assessment of the variability of diffusion metrics is essential for establishing the validity of using these quantitative metrics in multicenter trials. The low variability in ADC values across vendors and institutions and validates the use of ADC as a quantitative tumor marker in pediatric multicenter trials.

AB - Rationale and Objectives: Magnetic resonance diffusion imaging can characterize physiologic characteristics of pediatric brain tumors used to assess therapy response. The purpose of this study was to assess the variability of the apparent diffusion coefficient (ADC) along z-axis of scanners in the multicenter Pediatric Brain Tumor Consortium (PBTC). Materials and Methods: Ice-water diffusion phantoms for each PBTC site were distributed with a specific diffusion imaging protocol. The phantom was scanned four successive times to 1) confirm water in the tube reached thermal equilibrium and 2) allow for assessment of intra-examination ADC repeatability. ADC profiles across slice positions for each vendor and institution combination were characterized using linear regression modeling with a quadratic fit. Results: Eleven sites collected data with a high degree of compliance to the diffusion protocol for each scanner. The mean ADC value at slice position zero for vendor A was 1.123×10-3mm2/s, vendor B was 1.0964×10-3mm2/s, and vendor C was 1.110×10-3mm2/s. The percentage coefficient of variation across all sites was 0.309% (standard deviation=0.322). The ADC values conformed well to a second-order polynomial along the z-axis, (ie, following a linear model pattern with quadratic fit) for vendor-institution combinations and across vendor-institution combinations as shown in the longitudinal model. Conclusions: Assessment of the variability of diffusion metrics is essential for establishing the validity of using these quantitative metrics in multicenter trials. The low variability in ADC values across vendors and institutions and validates the use of ADC as a quantitative tumor marker in pediatric multicenter trials.

UR - http://www.scopus.com/inward/record.url?scp=84927638124&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84927638124&partnerID=8YFLogxK

U2 - 10.1016/j.acra.2014.10.006

DO - 10.1016/j.acra.2014.10.006

M3 - Article

VL - 22

SP - 363

EP - 369

JO - Academic Radiology

JF - Academic Radiology

SN - 1076-6332

IS - 3

ER -