2374 - Innovative Patient CT Scan Extensions for Accurate Organ Dose Estimations Following Proton and Photon Therapy: A Pilot Study

S. Morato Rafet¹, K. T. Griffin¹, J. Shin², C. Lee¹, and M. Mille¹; ¹National Cancer Institute, National Institutes of Health, Rockville, MD, ²Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Rockville, MD

Purpose/Objective(s): The Pediatric Proton and Photon Therapy Comparison Cohort is a multi-center retrospective cohort initiated to investigate subsequent cancer after pediatric radiotherapy. The study involves the collection of data for 10,000 proton and 10,000 photon therapy patients treated between 2007 and 2022 from pediatric radiotherapy institutions across the U.S. and Canada. Patient DICOM CT images are being collected when available; however, the planning CT typically only covers the region of the body near the target volume. We previously developed a method to extend partial body CT images called Anatomically Predictive Extension (APE). The purpose of this study is to demonstrate the application of APE to selected patients from various centers in the Pediatric Proton and Photon Therapy Comparison Cohort study. Materials/

Methods: Five patients were selected from each of the thirteen pediatric radiotherapy institutions currently participating in the study. We collected the CT images sets and used the auto-segmentation AI tool called Total Segmentator to obtain organ volumes. The method compares the skeleton of the original patient with the skeletons from a library containing 359 pediatric patients. Image registration is carried out by comparing skeletal images obtained through thresholding, accounting for transformations such as scaling, rotation, and translation.
Results: The APE method was applied to the selected patients, resulting in an extended CT and associated structure file containing up to 117 different organs or bone structures. The output comprises new CT image sets for the extended patient version in DICOM format and combined organ volumes or bone structures, available in either an RT structure DICOM file or a NIFTI format file. Subsequently, these output files will be used for organ dose estimations using the planning tool codes by simulating the radiation therapy treatments described in the cohort patient’s data repository.
Conclusion: This pilot study constitutes a crucial initial step in implementing the extension method within the Pediatric Proton and Photon Therapy Comparison Cohort. Following this study, efforts will be concentrated on addressing the challenge of enhancing the automation of the method for thousands of patients. This study is vital for improving our understanding of organ dose tolerances for pediatric patients undergoing proton or photon radiotherapy treatments.