Exploration of temporal stability and prognostic power of radiomic features based on electronic portal imaging device images

Mazen Soufi, Hidetaka Arimura, Takahiro Nakamoto, Taka aki Hirose, Ohga Saiji, Yoshiyuki Umezu, Hiroshi Honda, Tomonari Sasaki

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Purpose: We aimed to explore the temporal stability of radiomic features in the presence of tumor motion and the prognostic powers of temporally stable features. Methods: We selected single fraction dynamic electronic portal imaging device (EPID) (n = 275 frames) and static digitally reconstructed radiographs (DRRs) of 11 lung cancer patients, who received stereotactic body radiation therapy (SBRT) under free breathing. Forty-seven statistical radiomic features, which consisted of 14 histogram-based features and 33 texture features derived from the graylevel co-occurrence and graylevel run-length matrices, were computed. The temporal stability was assessed by using a multiplication of the intra-class correlation coefficients (ICCs) between features derived from the EPID and DRR images at three quantization levels. The prognostic powers of the features were investigated using a different database of lung cancer patients (n = 221) based on a Kaplan-Meier survival analysis. Results: Fifteen radiomic features were found to be temporally stable for various quantization levels. Among these features, seven features have shown potentials for prognostic prediction in lung cancer patients. Conclusions: This study suggests a novel approach to select temporally stable radiomic features, which could hold prognostic powers in lung cancer patients.

Original languageEnglish
Pages (from-to)32-44
Number of pages13
JournalPhysica Medica
Volume46
DOIs
Publication statusPublished - Feb 1 2018

Fingerprint

lungs
Lung Neoplasms
cancer
Equipment and Supplies
electronics
Kaplan-Meier Estimate
Survival Analysis
breathing
multiplication
histograms
correlation coefficients
radiation therapy
Respiration
Radiotherapy
tumors
textures
occurrences
Databases
matrices
predictions

All Science Journal Classification (ASJC) codes

  • Biophysics
  • Radiology Nuclear Medicine and imaging
  • Physics and Astronomy(all)

Cite this

Exploration of temporal stability and prognostic power of radiomic features based on electronic portal imaging device images. / Soufi, Mazen; Arimura, Hidetaka; Nakamoto, Takahiro; Hirose, Taka aki; Saiji, Ohga; Umezu, Yoshiyuki; Honda, Hiroshi; Sasaki, Tomonari.

In: Physica Medica, Vol. 46, 01.02.2018, p. 32-44.

Research output: Contribution to journalArticle

Soufi, Mazen ; Arimura, Hidetaka ; Nakamoto, Takahiro ; Hirose, Taka aki ; Saiji, Ohga ; Umezu, Yoshiyuki ; Honda, Hiroshi ; Sasaki, Tomonari. / Exploration of temporal stability and prognostic power of radiomic features based on electronic portal imaging device images. In: Physica Medica. 2018 ; Vol. 46. pp. 32-44.
@article{4384141c209f4df594933f69f2d6d5cf,
title = "Exploration of temporal stability and prognostic power of radiomic features based on electronic portal imaging device images",
abstract = "Purpose: We aimed to explore the temporal stability of radiomic features in the presence of tumor motion and the prognostic powers of temporally stable features. Methods: We selected single fraction dynamic electronic portal imaging device (EPID) (n = 275 frames) and static digitally reconstructed radiographs (DRRs) of 11 lung cancer patients, who received stereotactic body radiation therapy (SBRT) under free breathing. Forty-seven statistical radiomic features, which consisted of 14 histogram-based features and 33 texture features derived from the graylevel co-occurrence and graylevel run-length matrices, were computed. The temporal stability was assessed by using a multiplication of the intra-class correlation coefficients (ICCs) between features derived from the EPID and DRR images at three quantization levels. The prognostic powers of the features were investigated using a different database of lung cancer patients (n = 221) based on a Kaplan-Meier survival analysis. Results: Fifteen radiomic features were found to be temporally stable for various quantization levels. Among these features, seven features have shown potentials for prognostic prediction in lung cancer patients. Conclusions: This study suggests a novel approach to select temporally stable radiomic features, which could hold prognostic powers in lung cancer patients.",
author = "Mazen Soufi and Hidetaka Arimura and Takahiro Nakamoto and Hirose, {Taka aki} and Ohga Saiji and Yoshiyuki Umezu and Hiroshi Honda and Tomonari Sasaki",
year = "2018",
month = "2",
day = "1",
doi = "10.1016/j.ejmp.2017.11.037",
language = "English",
volume = "46",
pages = "32--44",
journal = "Physica Medica",
issn = "1120-1797",
publisher = "Associazione Italiana di Fisica Medica",

}

TY - JOUR

T1 - Exploration of temporal stability and prognostic power of radiomic features based on electronic portal imaging device images

AU - Soufi, Mazen

AU - Arimura, Hidetaka

AU - Nakamoto, Takahiro

AU - Hirose, Taka aki

AU - Saiji, Ohga

AU - Umezu, Yoshiyuki

AU - Honda, Hiroshi

AU - Sasaki, Tomonari

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Purpose: We aimed to explore the temporal stability of radiomic features in the presence of tumor motion and the prognostic powers of temporally stable features. Methods: We selected single fraction dynamic electronic portal imaging device (EPID) (n = 275 frames) and static digitally reconstructed radiographs (DRRs) of 11 lung cancer patients, who received stereotactic body radiation therapy (SBRT) under free breathing. Forty-seven statistical radiomic features, which consisted of 14 histogram-based features and 33 texture features derived from the graylevel co-occurrence and graylevel run-length matrices, were computed. The temporal stability was assessed by using a multiplication of the intra-class correlation coefficients (ICCs) between features derived from the EPID and DRR images at three quantization levels. The prognostic powers of the features were investigated using a different database of lung cancer patients (n = 221) based on a Kaplan-Meier survival analysis. Results: Fifteen radiomic features were found to be temporally stable for various quantization levels. Among these features, seven features have shown potentials for prognostic prediction in lung cancer patients. Conclusions: This study suggests a novel approach to select temporally stable radiomic features, which could hold prognostic powers in lung cancer patients.

AB - Purpose: We aimed to explore the temporal stability of radiomic features in the presence of tumor motion and the prognostic powers of temporally stable features. Methods: We selected single fraction dynamic electronic portal imaging device (EPID) (n = 275 frames) and static digitally reconstructed radiographs (DRRs) of 11 lung cancer patients, who received stereotactic body radiation therapy (SBRT) under free breathing. Forty-seven statistical radiomic features, which consisted of 14 histogram-based features and 33 texture features derived from the graylevel co-occurrence and graylevel run-length matrices, were computed. The temporal stability was assessed by using a multiplication of the intra-class correlation coefficients (ICCs) between features derived from the EPID and DRR images at three quantization levels. The prognostic powers of the features were investigated using a different database of lung cancer patients (n = 221) based on a Kaplan-Meier survival analysis. Results: Fifteen radiomic features were found to be temporally stable for various quantization levels. Among these features, seven features have shown potentials for prognostic prediction in lung cancer patients. Conclusions: This study suggests a novel approach to select temporally stable radiomic features, which could hold prognostic powers in lung cancer patients.

UR - http://www.scopus.com/inward/record.url?scp=85043230412&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043230412&partnerID=8YFLogxK

U2 - 10.1016/j.ejmp.2017.11.037

DO - 10.1016/j.ejmp.2017.11.037

M3 - Article

C2 - 29519407

AN - SCOPUS:85043230412

VL - 46

SP - 32

EP - 44

JO - Physica Medica

JF - Physica Medica

SN - 1120-1797

ER -