Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies

Rachel B. Ger, Shouhao Zhou, Pai Chun Melinda Chi, Hannah J. Lee, Rick R. Layman, A. Kyle Jones, David L. Goff, Clifton D. Fuller, Rebecca M. Howell, Heng Li, R. Jason Stafford, Laurence E. Court, Dennis S. Mackin

Research output: Contribution to journalArticle

Abstract

Radiomics has shown promise in improving models for predicting patient outcomes. However, to maximize the information gain of the radiomics features, especially in larger patient cohorts, the variability in radiomics features owing to differences between scanners and scanning protocols must be accounted for. To this aim, the imaging variability of radiomics feature values was evaluated on 100 computed tomography scanners at 35 clinics by imaging a radiomics phantom using a controlled protocol and the commonly used chest and head protocols of the local clinic. We used a linear mixed-effects model to determine the degree to which the manufacturer and individual scanners contribute to the overall variability. Using a controlled protocol reduced the overall variability by 57% and 52% compared to the local chest and head protocols respectively. The controlled protocol also reduced the relative contribution of the manufacturer to the total variability. For almost all variabilities (manufacturer, scanner, and residual with different preprocesssing), the controlled protocol scans had a significantly smaller variability than the local protocol scans did. For most radiomics features, the imaging variability was small relative to the inter-patient feature variability in non–small cell lung cancer and head and neck squamous cell carcinoma patient cohorts. From this study, we conclude that using controlled scans can reduce the variability in radiomics features, and our results demonstrate the importance of using controlled protocols in prospective radiomics studies.

Original languageEnglish (US)
Article number13047
JournalScientific reports
Volume8
Issue number1
DOIs
StatePublished - Dec 1 2018
Externally publishedYes

Fingerprint

Thorax
Head
X-Ray Computed Tomography Scanners
Head and Neck Neoplasms
Non-Small Cell Lung Carcinoma
Prospective Studies
Carcinoma, squamous cell of head and neck

ASJC Scopus subject areas

  • General

Cite this

Ger, R. B., Zhou, S., Chi, P. C. M., Lee, H. J., Layman, R. R., Jones, A. K., ... Mackin, D. S. (2018). Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies. Scientific reports, 8(1), [13047]. https://doi.org/10.1038/s41598-018-31509-z

Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies. / Ger, Rachel B.; Zhou, Shouhao; Chi, Pai Chun Melinda; Lee, Hannah J.; Layman, Rick R.; Jones, A. Kyle; Goff, David L.; Fuller, Clifton D.; Howell, Rebecca M.; Li, Heng; Stafford, R. Jason; Court, Laurence E.; Mackin, Dennis S.

In: Scientific reports, Vol. 8, No. 1, 13047, 01.12.2018.

Research output: Contribution to journalArticle

Ger, RB, Zhou, S, Chi, PCM, Lee, HJ, Layman, RR, Jones, AK, Goff, DL, Fuller, CD, Howell, RM, Li, H, Stafford, RJ, Court, LE & Mackin, DS 2018, 'Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies', Scientific reports, vol. 8, no. 1, 13047. https://doi.org/10.1038/s41598-018-31509-z
Ger, Rachel B. ; Zhou, Shouhao ; Chi, Pai Chun Melinda ; Lee, Hannah J. ; Layman, Rick R. ; Jones, A. Kyle ; Goff, David L. ; Fuller, Clifton D. ; Howell, Rebecca M. ; Li, Heng ; Stafford, R. Jason ; Court, Laurence E. ; Mackin, Dennis S. / Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies. In: Scientific reports. 2018 ; Vol. 8, No. 1.
@article{eb843de1c75b4bc1a089d13fd18fea1b,
title = "Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies",
abstract = "Radiomics has shown promise in improving models for predicting patient outcomes. However, to maximize the information gain of the radiomics features, especially in larger patient cohorts, the variability in radiomics features owing to differences between scanners and scanning protocols must be accounted for. To this aim, the imaging variability of radiomics feature values was evaluated on 100 computed tomography scanners at 35 clinics by imaging a radiomics phantom using a controlled protocol and the commonly used chest and head protocols of the local clinic. We used a linear mixed-effects model to determine the degree to which the manufacturer and individual scanners contribute to the overall variability. Using a controlled protocol reduced the overall variability by 57{\%} and 52{\%} compared to the local chest and head protocols respectively. The controlled protocol also reduced the relative contribution of the manufacturer to the total variability. For almost all variabilities (manufacturer, scanner, and residual with different preprocesssing), the controlled protocol scans had a significantly smaller variability than the local protocol scans did. For most radiomics features, the imaging variability was small relative to the inter-patient feature variability in non–small cell lung cancer and head and neck squamous cell carcinoma patient cohorts. From this study, we conclude that using controlled scans can reduce the variability in radiomics features, and our results demonstrate the importance of using controlled protocols in prospective radiomics studies.",
author = "Ger, {Rachel B.} and Shouhao Zhou and Chi, {Pai Chun Melinda} and Lee, {Hannah J.} and Layman, {Rick R.} and Jones, {A. Kyle} and Goff, {David L.} and Fuller, {Clifton D.} and Howell, {Rebecca M.} and Heng Li and Stafford, {R. Jason} and Court, {Laurence E.} and Mackin, {Dennis S.}",
year = "2018",
month = "12",
day = "1",
doi = "10.1038/s41598-018-31509-z",
language = "English (US)",
volume = "8",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies

AU - Ger, Rachel B.

AU - Zhou, Shouhao

AU - Chi, Pai Chun Melinda

AU - Lee, Hannah J.

AU - Layman, Rick R.

AU - Jones, A. Kyle

AU - Goff, David L.

AU - Fuller, Clifton D.

AU - Howell, Rebecca M.

AU - Li, Heng

AU - Stafford, R. Jason

AU - Court, Laurence E.

AU - Mackin, Dennis S.

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Radiomics has shown promise in improving models for predicting patient outcomes. However, to maximize the information gain of the radiomics features, especially in larger patient cohorts, the variability in radiomics features owing to differences between scanners and scanning protocols must be accounted for. To this aim, the imaging variability of radiomics feature values was evaluated on 100 computed tomography scanners at 35 clinics by imaging a radiomics phantom using a controlled protocol and the commonly used chest and head protocols of the local clinic. We used a linear mixed-effects model to determine the degree to which the manufacturer and individual scanners contribute to the overall variability. Using a controlled protocol reduced the overall variability by 57% and 52% compared to the local chest and head protocols respectively. The controlled protocol also reduced the relative contribution of the manufacturer to the total variability. For almost all variabilities (manufacturer, scanner, and residual with different preprocesssing), the controlled protocol scans had a significantly smaller variability than the local protocol scans did. For most radiomics features, the imaging variability was small relative to the inter-patient feature variability in non–small cell lung cancer and head and neck squamous cell carcinoma patient cohorts. From this study, we conclude that using controlled scans can reduce the variability in radiomics features, and our results demonstrate the importance of using controlled protocols in prospective radiomics studies.

AB - Radiomics has shown promise in improving models for predicting patient outcomes. However, to maximize the information gain of the radiomics features, especially in larger patient cohorts, the variability in radiomics features owing to differences between scanners and scanning protocols must be accounted for. To this aim, the imaging variability of radiomics feature values was evaluated on 100 computed tomography scanners at 35 clinics by imaging a radiomics phantom using a controlled protocol and the commonly used chest and head protocols of the local clinic. We used a linear mixed-effects model to determine the degree to which the manufacturer and individual scanners contribute to the overall variability. Using a controlled protocol reduced the overall variability by 57% and 52% compared to the local chest and head protocols respectively. The controlled protocol also reduced the relative contribution of the manufacturer to the total variability. For almost all variabilities (manufacturer, scanner, and residual with different preprocesssing), the controlled protocol scans had a significantly smaller variability than the local protocol scans did. For most radiomics features, the imaging variability was small relative to the inter-patient feature variability in non–small cell lung cancer and head and neck squamous cell carcinoma patient cohorts. From this study, we conclude that using controlled scans can reduce the variability in radiomics features, and our results demonstrate the importance of using controlled protocols in prospective radiomics studies.

UR - http://www.scopus.com/inward/record.url?scp=85052714792&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052714792&partnerID=8YFLogxK

U2 - 10.1038/s41598-018-31509-z

DO - 10.1038/s41598-018-31509-z

M3 - Article

VL - 8

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 13047

ER -