CA Owens^1,2*, B Rigaud³, E Ludmir^4,5, A Gupta^1,2, S Shrestha^1,2, AC Paulino⁴, SA Smith¹, CB Peterson⁵, SF Kry^1,2, C Lee⁶, T Henderson⁷, GT Armstrong⁸, KK Brock^1,3, RM Howell^1,2, (1) The University of Texas MD Anderson Cancer Center, Department of Radiation Physics, Houston, TX, USA, (2) MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences, Graduate Program in Medical Physics, Houston, TX USA, (3) The University of Texas MD Anderson Cancer Center, Department of Imaging Physics, Houston, TX, USA, (4) The University of Texas MD Anderson Cancer Center, Department of Radiation Oncology, Houston, TX, USA, (5) The University of Texas MD Anderson Cancer Center, Department of Biostatistics, Houston, TX, USA, (6) National Cancer Institute, Division of Cancer Epidemiology and Genetics, Bethesda, MD, USA, (7) The University of Chicago, Department of Pediatrics, Chicago, IL, USA, (8) St. Jude Childrens Research Hospital, Department of Epidemiology and Cancer Control, Memphis, TN, USA

MO-FG-BRB-5 (Monday, 7/11/2022) 1:45 PM - 3:45 PM [Eastern Time (GMT-4)]

Ballroom B

Purpose: Study purposes were to develop and integrate a colorectal model that incorporates anatomical variations of pediatric patients into an age-scalable computational phantom, and to validate the model for pediatric radiation therapy (RT) dose reconstructions.

Methods: Colorectal contours were manually derived by two physicians from whole-body non-contrast CT scans of 114 pediatric patients (age range: 2.1-21.6years, 74males, 40females). One contour was used for an anatomical template, 103 for training and 10 for testing. Training contours were used to create a colorectal principal component analysis (PCA)-based statistical shape model (SSM) to extract the population’s dominant deformations. The SSM was integrated into our in-house age-scalable phantom. Geometric accuracy was assessed between patient-specific and SSM reconstructed contours. Two alternative colorectal shapes (hereafter alternative #1, #2) were generated using the first 17 dominant SSM modes. Dosimetric accuracy was assessed by comparing colorectal doses from 10 test patients’ CT-based RT plans (ground-truth) with reconstructed doses for the mean, alternative #1 and #2 colorectal models in age-matched phantoms. Lastly, a proof-of-concept study was conducted to demonstrate that our colorectal models can be integrated into any computational phantom.

Results: Using all 103 PCA modes, mean(min-max) Dice similarity coefficient, distance-to-agreement and Hausdorff distance between patient-specific and reconstructed contours were 0.89(0.85-0.91), 2.1mm(1.7-3.0), and 8.6mm(5.7-14.3), respectively. Average absolute difference between ground-truth and reconstructed mean and maximum colorectal doses (normalized to 20Gy prescription dose) for the mean colorectal model (alternative #1, alternative #2) were 6.33%(8.08%, 6.13%) and 4.38%(4.28%, 4.65%), respectively. Similar agreement was observed when colorectal models were integrated into a collaborator’s 5-year-old phantom; 6.28%(11.73%, 0.34%) and 0.47%(0.19%, 0.90%), respectively.

Conclusion: We developed and validated a population-based colorectal SSM and demonstrated its use for pediatric RT dose reconstruction in two phantoms. We will use this SSM to reconstruct pre-CT era colorectal doses for irradiated individuals in the Childhood Cancer Survivor Study.

Funding Support, Disclosures, and Conflict of Interest: This work was supported by the National Cancer Institute (CA55727, G.T. Armstrong, Principal Investigator).

Keywords

Phantoms, Radiation Dosimetry, Radiation Effects

Taxonomy

TH- Response Assessment: Radiation induced second cancers

Contact Email