Radiation Physics

PV QA 3 - Poster Viewing Q&A 3

TU_20_3314 - Radiomics Analysis of Normal Tissue for Patients with Lung Cancers

Tuesday, October 23
1:00 PM - 2:30 PM
Location: Innovation Hub, Exhibit Hall 3

Radiomics Analysis of Normal Tissue for Patients with Lung Cancers
H. Bagher-Ebadian, Q. Wu, A. I. Ghanem, C. Liu, S. L. Brown, N. Wen, M. Ajlouni, M. Simoff, B. Movsas, and I. J. Chetty; Henry Ford Health System, Detroit, MI

Purpose/Objective(s): To perform radiomics analyses on normal lung tissue delineated from CT-based image datasets for patients with locally advanced, non-small cell lung cancers (NSCLC) in order to characterize differences between patients with and without radiation-induced pneumonitis (RP).

Materials/Methods: Planning and 3-month follow-up CT image datasets of forty-one patients (14 with RP and 27 with no evidence of RP) with stage-III lung cancers, treated with IMRT/3D-CRT, were investigated. One hundred sixty-eight radiomics features were extracted from the volume of normal lung tissue receiving ≥20 Gy, excluding the ITV, according to the following 8 different classes: Intensity Histogram Based Features (IHBF), Gray Level Run Length (GLRL), Law’s Textural information (LAWS), Discrete Orthonormal Stockwell Transform (DOST), Local Binary Pattern (LBP), Two-Dimensional Wavelet Transform (2DWT), Two Dimensional Gabor Filter (2DGF), and Gray Level Co-Occurrence Matrix (GLCM). Analysis of variance (ANOVA) was used to compare differences between radiomics features. Fisher's-combined method with mean-percent-difference (MPD) measures was used to compare CT-based radiomics features between non-RP and RP patients.

Results: Among 168 radiomics features compared on planning CT images, 2 were significantly different between RP and non-RP groups: Intensity-Based-Histogram-Feature (IBHF, entropy): p=0.0334, MPD=-27.71% and 2D-Wavelet-Transform (2DWT, entropy): p=0.0214, and MPD= 185.59%. For patients with RP, 6 features were significantly different, with pFisher<0.001 for all categories: IBHF, DOST, and DTW, between RP grades 1 and 2 sub-groups. On follow-up CT images, 62 features were significantly different between RP and non-RP groups, with pFisher<0.001, in all categories: IBHF (MPD=57.78%), Gray-Level-Run-Length (MPD=28.61%), LAWS (MPD=-15.74%), Discrete-Orthonormal-Stockwell-Transform (MPD=13.37%), Local-Binary-Pattern (MPD=30.33%), 2D-Wavelet-Transform (MPD=37.02%), 2D-Gabor-Filter (MPD=-3.08%), and Gray-Level-Co-occurrence-Matrix: (MPD=-30.83%). Eighteen features were significantly different between RP grades 1 and 2 sub-groups, with pFisher<0.001 in six categories: IBHF, GLRL, LAWS, DOST, 2DWT, GLCM.

Conclusion: Results suggest that the entropy of normal lung tissue, associated with the arrangement of intensities (IBHF) and frequency homogeneity (2DWT) on planning CT datasets demonstrate promising biomarkers for development of radiation-induced pneumonitis. High-frequency image components were found to be significantly impacted by RT between RP and non-RP CT datasets. The results of this pilot study, albeit subject to confirmation in a larger patient population, suggest a potential role for the use of radiomics-based signatures in models developed for predicting radiation-induced lung damage in patients with locally advanced-NSCLC.

Author Disclosure: H. Bagher-Ebadian: None. Q. Wu: None. A.I. Ghanem: None. C. Liu: None. S.L. Brown: None. N. Wen: Research Grant; American Cancer Society. M. Simoff: Advisory Board; Varian Medical. President; Association of Interventional Pulmonology Fellowsh. B. Movsas: Research Grant; Varian Medical Systems, Inc, Philips, Inc, NCI. Travel Expenses; Varian Medical Systems, Inc, Philips, Inc. Patent/License Fees/Copyright; Henry Ford Health System. Chair, Quality of Life Subcommittee; Radiation Therapy Oncology Group/NRG. I.J. Chetty: Research Grant; Varian Medical Systems, Inc, Philips Healthcare. Travel Expenses; Varian Medical Systems, Inc. ASTRO Nominating Committee member; ASTRO Nominating Committee.

Hassan Bagher-Ebadian, PhD

Henry Ford Health System

Presentation(s):

Send Email for Hassan Bagher-Ebadian


Assets

TU_20_3314 - Radiomics Analysis of Normal Tissue for Patients with Lung Cancers



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Radiomics Analysis of Normal Tissue for Patients with Lung Cancers