Incorporation of pre-therapy 18F-FDG uptake data with CT texture features into a radiomics model for radiation pneumonitis diagnosis

Gregory J. Anthony, Alexandra Cunliffe, Richard Castillo, Ngoc Pham, Thomas Guerrero, Samuel G. Armato, Hania A. Al-Hallaq

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Purpose: To determine whether the addition of standardized uptake value (SUV) from PET scans to CT lung texture features could improve a radiomics-based model of radiation pneumonitis (RP) diagnosis in patients undergoing radiotherapy. Methods and materials: Anonymized data from 96 esophageal cancer patients (18 RP-positive cases of Grade ≥ 2) were collected including pre-therapy PET/CT scans, pre-/post-therapy diagnostic CT scans and RP status. Twenty texture features (first-order, fractal, Laws' filter and gray-level co-occurrence matrix) were calculated from diagnostic CT scans and compared in anatomically matched regions of the lung. Classifier performance (texture, SUV, or combination) was assessed by calculating the area under the receiver operating characteristic curve (AUC). For each texture feature, logistic regression classifiers consisting of the average change in texture feature value and the pre-therapy SUV standard deviation (SUVSD) were created and compared with the texture feature as a lone classifier using ANOVA with correction for multiple comparisons (P < 0.0025). Results: While clinical parameters (mean lung dose, smoking history, tumor location) were not significantly different among patients with and without symptomatic RP, SUV and texture parameters were significantly associated with RP status. AUC for single-texture feature classifiers alone ranged from 0.58 to 0.81 and 0.53 to 0.71 in high-dose (≥ 30 Gy) and low-dose (< 10 Gy) regions of the lungs, respectively. AUC for SUVSD alone was 0.69 (95% confidence interval: 0.54-0.83). Adding SUVSD into a logistic regression model significantly improved model fit for 18, 14 and 11 texture features and increased the mean AUC across features by 0.08, 0.06, and 0.04 in the low-, medium-, and high-dose regions, respectively. Conclusions: Addition of SUVSD to a single-texture feature improves classifier performance on average, but the improvement is smaller in magnitude when SUVSD is added to an already effective classifier using texture alone. These findings demonstrate the potential for more accurate assessment of RP using information from multiple imaging modalities.

Original languageEnglish (US)
Pages (from-to)3686-3694
Number of pages9
JournalMedical Physics
Volume44
Issue number7
DOIs
StatePublished - Jul 1 2017

Fingerprint

Radiation Pneumonitis
Fluorodeoxyglucose F18
Area Under Curve
Lung
Logistic Models
Therapeutics
Fractals
Esophageal Neoplasms
ROC Curve
Analysis of Variance
Radiotherapy
Smoking
History
Confidence Intervals
Neoplasms

Keywords

  • CT
  • PET
  • radiation pneumonitis
  • radiomics
  • SUV
  • texture analysis

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

Anthony, G. J., Cunliffe, A., Castillo, R., Pham, N., Guerrero, T., Armato, S. G., & Al-Hallaq, H. A. (2017). Incorporation of pre-therapy 18F-FDG uptake data with CT texture features into a radiomics model for radiation pneumonitis diagnosis. Medical Physics, 44(7), 3686-3694. https://doi.org/10.1002/mp.12282

Incorporation of pre-therapy 18F-FDG uptake data with CT texture features into a radiomics model for radiation pneumonitis diagnosis. / Anthony, Gregory J.; Cunliffe, Alexandra; Castillo, Richard; Pham, Ngoc; Guerrero, Thomas; Armato, Samuel G.; Al-Hallaq, Hania A.

In: Medical Physics, Vol. 44, No. 7, 01.07.2017, p. 3686-3694.

Research output: Contribution to journalArticle

Anthony, GJ, Cunliffe, A, Castillo, R, Pham, N, Guerrero, T, Armato, SG & Al-Hallaq, HA 2017, 'Incorporation of pre-therapy 18F-FDG uptake data with CT texture features into a radiomics model for radiation pneumonitis diagnosis', Medical Physics, vol. 44, no. 7, pp. 3686-3694. https://doi.org/10.1002/mp.12282
Anthony, Gregory J. ; Cunliffe, Alexandra ; Castillo, Richard ; Pham, Ngoc ; Guerrero, Thomas ; Armato, Samuel G. ; Al-Hallaq, Hania A. / Incorporation of pre-therapy 18F-FDG uptake data with CT texture features into a radiomics model for radiation pneumonitis diagnosis. In: Medical Physics. 2017 ; Vol. 44, No. 7. pp. 3686-3694.
@article{189e4d4d3d024ccb867ae2b47c352844,
title = "Incorporation of pre-therapy 18F-FDG uptake data with CT texture features into a radiomics model for radiation pneumonitis diagnosis",
abstract = "Purpose: To determine whether the addition of standardized uptake value (SUV) from PET scans to CT lung texture features could improve a radiomics-based model of radiation pneumonitis (RP) diagnosis in patients undergoing radiotherapy. Methods and materials: Anonymized data from 96 esophageal cancer patients (18 RP-positive cases of Grade ≥ 2) were collected including pre-therapy PET/CT scans, pre-/post-therapy diagnostic CT scans and RP status. Twenty texture features (first-order, fractal, Laws' filter and gray-level co-occurrence matrix) were calculated from diagnostic CT scans and compared in anatomically matched regions of the lung. Classifier performance (texture, SUV, or combination) was assessed by calculating the area under the receiver operating characteristic curve (AUC). For each texture feature, logistic regression classifiers consisting of the average change in texture feature value and the pre-therapy SUV standard deviation (SUVSD) were created and compared with the texture feature as a lone classifier using ANOVA with correction for multiple comparisons (P < 0.0025). Results: While clinical parameters (mean lung dose, smoking history, tumor location) were not significantly different among patients with and without symptomatic RP, SUV and texture parameters were significantly associated with RP status. AUC for single-texture feature classifiers alone ranged from 0.58 to 0.81 and 0.53 to 0.71 in high-dose (≥ 30 Gy) and low-dose (< 10 Gy) regions of the lungs, respectively. AUC for SUVSD alone was 0.69 (95{\%} confidence interval: 0.54-0.83). Adding SUVSD into a logistic regression model significantly improved model fit for 18, 14 and 11 texture features and increased the mean AUC across features by 0.08, 0.06, and 0.04 in the low-, medium-, and high-dose regions, respectively. Conclusions: Addition of SUVSD to a single-texture feature improves classifier performance on average, but the improvement is smaller in magnitude when SUVSD is added to an already effective classifier using texture alone. These findings demonstrate the potential for more accurate assessment of RP using information from multiple imaging modalities.",
keywords = "CT, PET, radiation pneumonitis, radiomics, SUV, texture analysis",
author = "Anthony, {Gregory J.} and Alexandra Cunliffe and Richard Castillo and Ngoc Pham and Thomas Guerrero and Armato, {Samuel G.} and Al-Hallaq, {Hania A.}",
year = "2017",
month = "7",
day = "1",
doi = "10.1002/mp.12282",
language = "English (US)",
volume = "44",
pages = "3686--3694",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "7",

}

TY - JOUR

T1 - Incorporation of pre-therapy 18F-FDG uptake data with CT texture features into a radiomics model for radiation pneumonitis diagnosis

AU - Anthony, Gregory J.

AU - Cunliffe, Alexandra

AU - Castillo, Richard

AU - Pham, Ngoc

AU - Guerrero, Thomas

AU - Armato, Samuel G.

AU - Al-Hallaq, Hania A.

PY - 2017/7/1

Y1 - 2017/7/1

N2 - Purpose: To determine whether the addition of standardized uptake value (SUV) from PET scans to CT lung texture features could improve a radiomics-based model of radiation pneumonitis (RP) diagnosis in patients undergoing radiotherapy. Methods and materials: Anonymized data from 96 esophageal cancer patients (18 RP-positive cases of Grade ≥ 2) were collected including pre-therapy PET/CT scans, pre-/post-therapy diagnostic CT scans and RP status. Twenty texture features (first-order, fractal, Laws' filter and gray-level co-occurrence matrix) were calculated from diagnostic CT scans and compared in anatomically matched regions of the lung. Classifier performance (texture, SUV, or combination) was assessed by calculating the area under the receiver operating characteristic curve (AUC). For each texture feature, logistic regression classifiers consisting of the average change in texture feature value and the pre-therapy SUV standard deviation (SUVSD) were created and compared with the texture feature as a lone classifier using ANOVA with correction for multiple comparisons (P < 0.0025). Results: While clinical parameters (mean lung dose, smoking history, tumor location) were not significantly different among patients with and without symptomatic RP, SUV and texture parameters were significantly associated with RP status. AUC for single-texture feature classifiers alone ranged from 0.58 to 0.81 and 0.53 to 0.71 in high-dose (≥ 30 Gy) and low-dose (< 10 Gy) regions of the lungs, respectively. AUC for SUVSD alone was 0.69 (95% confidence interval: 0.54-0.83). Adding SUVSD into a logistic regression model significantly improved model fit for 18, 14 and 11 texture features and increased the mean AUC across features by 0.08, 0.06, and 0.04 in the low-, medium-, and high-dose regions, respectively. Conclusions: Addition of SUVSD to a single-texture feature improves classifier performance on average, but the improvement is smaller in magnitude when SUVSD is added to an already effective classifier using texture alone. These findings demonstrate the potential for more accurate assessment of RP using information from multiple imaging modalities.

AB - Purpose: To determine whether the addition of standardized uptake value (SUV) from PET scans to CT lung texture features could improve a radiomics-based model of radiation pneumonitis (RP) diagnosis in patients undergoing radiotherapy. Methods and materials: Anonymized data from 96 esophageal cancer patients (18 RP-positive cases of Grade ≥ 2) were collected including pre-therapy PET/CT scans, pre-/post-therapy diagnostic CT scans and RP status. Twenty texture features (first-order, fractal, Laws' filter and gray-level co-occurrence matrix) were calculated from diagnostic CT scans and compared in anatomically matched regions of the lung. Classifier performance (texture, SUV, or combination) was assessed by calculating the area under the receiver operating characteristic curve (AUC). For each texture feature, logistic regression classifiers consisting of the average change in texture feature value and the pre-therapy SUV standard deviation (SUVSD) were created and compared with the texture feature as a lone classifier using ANOVA with correction for multiple comparisons (P < 0.0025). Results: While clinical parameters (mean lung dose, smoking history, tumor location) were not significantly different among patients with and without symptomatic RP, SUV and texture parameters were significantly associated with RP status. AUC for single-texture feature classifiers alone ranged from 0.58 to 0.81 and 0.53 to 0.71 in high-dose (≥ 30 Gy) and low-dose (< 10 Gy) regions of the lungs, respectively. AUC for SUVSD alone was 0.69 (95% confidence interval: 0.54-0.83). Adding SUVSD into a logistic regression model significantly improved model fit for 18, 14 and 11 texture features and increased the mean AUC across features by 0.08, 0.06, and 0.04 in the low-, medium-, and high-dose regions, respectively. Conclusions: Addition of SUVSD to a single-texture feature improves classifier performance on average, but the improvement is smaller in magnitude when SUVSD is added to an already effective classifier using texture alone. These findings demonstrate the potential for more accurate assessment of RP using information from multiple imaging modalities.

KW - CT

KW - PET

KW - radiation pneumonitis

KW - radiomics

KW - SUV

KW - texture analysis

UR - http://www.scopus.com/inward/record.url?scp=85021975128&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85021975128&partnerID=8YFLogxK

U2 - 10.1002/mp.12282

DO - 10.1002/mp.12282

M3 - Article

AN - SCOPUS:85021975128

VL - 44

SP - 3686

EP - 3694

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 7

ER -