skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: SU-F-R-38: Impact of Smoothing and Noise On Robustness of CBCT Textural Features for Prediction of Response to Radiotherapy Treatment of Head and Neck Cancers

Abstract

Purpose: To examine the impact of image smoothing and noise on the robustness of textural information extracted from CBCT images for prediction of radiotherapy response for patients with head/neck (H/N) cancers. Methods: CBCT image datasets for 14 patients with H/N cancer treated with radiation (70 Gy in 35 fractions) were investigated. A deformable registration algorithm was used to fuse planning CT’s to CBCT’s. Tumor volume was automatically segmented on each CBCT image dataset. Local control at 1-year was used to classify 8 patients as responders (R), and 6 as non-responders (NR). A smoothing filter [2D Adaptive Weiner (2DAW) with 3 different windows (ψ=3, 5, and 7)], and two noise models (Poisson and Gaussian, SNR=25) were implemented, and independently applied to CBCT images. Twenty-two textural features, describing the spatial arrangement of voxel intensities calculated from gray-level co-occurrence matrices, were extracted for all tumor volumes. Results: Relative to CBCT images without smoothing, none of 22 textural features extracted showed any significant differences when smoothing was applied (using the 2DAW with filtering parameters of ψ=3 and 5), in the responder and non-responder groups. When smoothing, 2DAW with ψ=7 was applied, one textural feature, Information Measure of Correlation, was significantly different relative to nomore » smoothing. Only 4 features (Energy, Entropy, Homogeneity, and Maximum-Probability) were found to be statistically different between the R and NR groups (Table 1). These features remained statistically significant discriminators for R and NR groups in presence of noise and smoothing. Conclusion: This preliminary work suggests that textural classifiers for response prediction, extracted from H&N CBCT images, are robust to low-power noise and low-pass filtering. While other types of filters will alter the spatial frequencies differently, these results are promising. The current study is subject to Type II errors. A much larger cohort of patients is needed to confirm these results. This work was supported in part by a grant from Varian Medical Systems (Palo Alto, CA)« less

Authors:
; ; ; ;  [1]
  1. Henry Ford Health System, Detroit, MI (United States)
Publication Date:
OSTI Identifier:
22626758
Resource Type:
Journal Article
Resource Relation:
Journal Name: Medical Physics; Journal Volume: 43; Journal Issue: 6; Other Information: (c) 2016 American Association of Physicists in Medicine; Country of input: International Atomic Energy Agency (IAEA)
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES; 61 RADIATION PROTECTION AND DOSIMETRY; ALGORITHMS; COMPUTERIZED TOMOGRAPHY; CORRELATIONS; DATASETS; DISCRIMINATORS; HEAD; IMAGES; NECK; NEOPLASMS; PATIENTS; RADIOTHERAPY

Citation Formats

Bagher-Ebadian, H, Chetty, I, Liu, C, Movsas, B, and Siddiqui, F. SU-F-R-38: Impact of Smoothing and Noise On Robustness of CBCT Textural Features for Prediction of Response to Radiotherapy Treatment of Head and Neck Cancers. United States: N. p., 2016. Web. doi:10.1118/1.4955809.
Bagher-Ebadian, H, Chetty, I, Liu, C, Movsas, B, & Siddiqui, F. SU-F-R-38: Impact of Smoothing and Noise On Robustness of CBCT Textural Features for Prediction of Response to Radiotherapy Treatment of Head and Neck Cancers. United States. doi:10.1118/1.4955809.
Bagher-Ebadian, H, Chetty, I, Liu, C, Movsas, B, and Siddiqui, F. 2016. "SU-F-R-38: Impact of Smoothing and Noise On Robustness of CBCT Textural Features for Prediction of Response to Radiotherapy Treatment of Head and Neck Cancers". United States. doi:10.1118/1.4955809.
@article{osti_22626758,
title = {SU-F-R-38: Impact of Smoothing and Noise On Robustness of CBCT Textural Features for Prediction of Response to Radiotherapy Treatment of Head and Neck Cancers},
author = {Bagher-Ebadian, H and Chetty, I and Liu, C and Movsas, B and Siddiqui, F},
abstractNote = {Purpose: To examine the impact of image smoothing and noise on the robustness of textural information extracted from CBCT images for prediction of radiotherapy response for patients with head/neck (H/N) cancers. Methods: CBCT image datasets for 14 patients with H/N cancer treated with radiation (70 Gy in 35 fractions) were investigated. A deformable registration algorithm was used to fuse planning CT’s to CBCT’s. Tumor volume was automatically segmented on each CBCT image dataset. Local control at 1-year was used to classify 8 patients as responders (R), and 6 as non-responders (NR). A smoothing filter [2D Adaptive Weiner (2DAW) with 3 different windows (ψ=3, 5, and 7)], and two noise models (Poisson and Gaussian, SNR=25) were implemented, and independently applied to CBCT images. Twenty-two textural features, describing the spatial arrangement of voxel intensities calculated from gray-level co-occurrence matrices, were extracted for all tumor volumes. Results: Relative to CBCT images without smoothing, none of 22 textural features extracted showed any significant differences when smoothing was applied (using the 2DAW with filtering parameters of ψ=3 and 5), in the responder and non-responder groups. When smoothing, 2DAW with ψ=7 was applied, one textural feature, Information Measure of Correlation, was significantly different relative to no smoothing. Only 4 features (Energy, Entropy, Homogeneity, and Maximum-Probability) were found to be statistically different between the R and NR groups (Table 1). These features remained statistically significant discriminators for R and NR groups in presence of noise and smoothing. Conclusion: This preliminary work suggests that textural classifiers for response prediction, extracted from H&N CBCT images, are robust to low-power noise and low-pass filtering. While other types of filters will alter the spatial frequencies differently, these results are promising. The current study is subject to Type II errors. A much larger cohort of patients is needed to confirm these results. This work was supported in part by a grant from Varian Medical Systems (Palo Alto, CA)},
doi = {10.1118/1.4955809},
journal = {Medical Physics},
number = 6,
volume = 43,
place = {United States},
year = 2016,
month = 6
}
  • Purpose: To evaluate whether tumor textural features extracted from both pre- and mid-treatment FDG-PET images predict early response to chemoradiotherapy in locally advanced head and neck cancer, and investigate whether they provide complementary value to conventional volume-based measurements. Methods: Ninety-four patients with locally advanced head and neck cancers were retrospectively studied. All patients received definitive chemoradiotherapy and underwent FDG-PET planning scans both before and during treatment. Within the primary tumor we extracted 6 textural features based on gray-level co-occurrence matrices (GLCM): entropy, dissimilarity, contrast, correlation, energy, and homogeneity. These image features were evaluated for their predictive power of treatment responsemore » to chemoradiotherapy in terms of local recurrence free survival (LRFS) and progression free survival (PFS). Logrank test were used to assess the statistical significance of the stratification between low- and high-risk groups. P-values were adjusted for multiple comparisons by the false discovery rate (FDR) method. Results: All six textural features extracted from pre-treatment PET images significantly differentiated low- and high-risk patient groups for LRFS (P=0.011–0.038) and PFS (P=0.029–0.034). On the other hand, none of the textural features on mid-treatment PET images was statistically significant in stratifying LRFS (P=0.212–0.445) or PFS (P=0.168–0.299). An imaging signature that combines textural feature (GLCM homogeneity) and metabolic tumor volume showed an improved performance for predicting LRFS (hazard ratio: 22.8, P<0.0001) and PFS (hazard ratio: 13.9, P=0.0005) in leave-one-out cross validation. Intra-tumor heterogeneity measured by textural features was significantly lower in mid-treatment PET images than in pre-treatment PET images (T-test: P<1.4e-6). Conclusion: Tumor textural features on pretreatment FDG-PET images are predictive for response to chemoradiotherapy in locally advanced head and neck cancer. The complementary information offered by textural features improves patient stratification and may potentially aid in personalized risk-adaptive therapy.« less
  • As part of the RTOG research effort in the treatment of advanced, inoperable squamous cancer of the head and neck region, the hypoxic cell sensitizer, misonidazole, was selected for investigation as an adjuvant to definitive irradiation. Based upon a pilot experience (78-02) showing a 67% complete response rate among 36 AJC Stage III-IV patients receiving full-dose irradiation and 6 weekly p.o. doses of misonidazole, a phase III trial was carried out from '79-'83. Three hundred and six patients were entered, 42% of whom had oropharyngeal primaries and with 78% of all cases representing T3 or T4 (inoperable) lesions. Only 16%more » of the entire series presented with N0 necks. Fractionation was altered among the misonidazole-receiving patients, in contrast to standard 5 treatments per week among control patients, such that 2 separate treatments were given on each day of p.o. misonidazole administration (2.0 gm/m2/wk X 6 doses, 2.5 Gy in a.m., 2.1 Gy in p.m.). Total tumor doses were identical among the two treatment arms except that a limitation of 40.0 Gy to spinal cord was specified for sensitized radiotherapy vs. 45.0 Gy for control patients. Primary tumor clearance was observed to be 55-60%, with minor variations according to tumor stage and site. The local regional control rate among radiotherapy-alone patients was 26% at 2 years compared to 22% (2 years) within the misonidazole-receiving group. Analysis of survival revealed no advantage to the sensitized patients, with 55 +/- 2% surviving 1 year and 22 +/- 1% living 3 years following treatment in both treatment categories. Distant metastases as first site of failure (12-13%) and the local failure among initial complete responders (46%) showed no advantage to the misonidazole group. Although a misonidazole dosage of 2.0 gm/m2/wk X 6 (12 gm/m2 total) is well tolerated, no clinical benefit was demonstrated in this randomized trial.« less
  • The risk of a second primary cancer arising in the head and neck, following surgical or radiation treatment of an initial primary cancer in the head and neck, was evaluated for 2,151 patients whose first cancers were diagnosed and treated at UCLA between 1955 and 1979. Based on follow-up data ranging from 5 to 30 years, the rate of development of second cancers of the head and neck was in excess of 2.5 per 1000 person-years at risk. There was no statistically significant difference in the frequency or post-treatment interval of second primary cancers related to the type of treatmentmore » of the first cancer, whether that was surgery, radiation therapy, or surgery plus radiation therapy.« less
  • Purpose: To evaluate the usefulness of a six-degrees-of freedom (6D) correction using ExacTrac robotics system in patients with head-and-neck (HN) cancer receiving radiation therapy.Methods: Local setup accuracy was analyzed for 12 patients undergoing intensity-modulated radiation therapy (IMRT). Patient position was imaged daily upon two different protocols, cone-beam computed tomography (CBCT), and ExacTrac (ET) images correction. Setup data from either approach were compared in terms of both residual errors after correction and punctual displacement of selected regions of interest (Mandible, C2, and C6 vertebral bodies).Results: On average, both protocols achieved reasonably low residual errors after initial correction. The observed differences inmore » shift vectors between the two protocols showed that CBCT tends to weight more C2 and C6 at the expense of the mandible, while ET tends to average more differences among the different ROIs.Conclusions: CBCT, even without 6D correction capabilities, seems preferable to ET for better consistent alignment and the capability to see soft tissues. Therefore, in our experience, CBCT represents a benchmark for positioning head and neck cancer patients.« less