ESTRO 2024 - Abstract Book

S3093

Physics - Autosegmentation

ESTRO 2024

Deep-learning-based autosegmentation was successful in deriving comparable contours between pre-treatment and subsequent simulation and diagnostic scans, including lower-resolution PET/CT scans. Using definitive radiotherapy for oropharyngeal cancer and resultant risk of carotid stenosis as a pilot, a non-statistically significant trend against the frequency of stenosis was observed for IMPT when compared with IMRT, and further evaluation is required to characterize this trend. Specific to the carotids, pre-treatment calcification ratio emerged as a convenient, rapid means of assessing patients' risk of carotid stenosis that is available before treatment. Mean carotid area did not appear to correlate to development of stenosis, but may correlate to dose delivered to carotids; additional dose levels are required to evaluate this. Utilizing autosegmentation on diagnostic scans for evaluation of volume changes over time appears feasible with careful analysis. This method is ready for expansion in larger patient cohorts as well as in other organs at risk for longitudinal evaluation of structures via already-attained diagnostic imaging.

Keywords: proton, carotid stenosis, artificial intelligence

References:

Neurol. 2014;2:183-92.

[2] Saxena A, Ng EYK, Lim ST. Imaging modalities to diagnose carotid artery stenosis: progress and prospect. Biomed Eng Online. 2019;18:66.

[3] Leclerc X, Godefroy O, Lucas C, Benhaim JF, Michel TS, Leys D, et al. Internal carotid arterial stenosis: CT angiography with volume rendering. Radiology. 1999;210:673-82.

[4] Samarzija K, Milosevic P, Jurjevic Z, Erdeljac E. Grading of carotid artery stenosis with computed tomography angiography: whether to use the narrowest diameter or the cross-sectional area. Insights Imaging. 2018;9:527-34.

2162

Digital Poster

Accurate Delineation of DIL via Boundary Enhancement for MRI-Guided Prostate SBRT

Xiao Liu 1 , Zirong Li 1 , Qichao Zhou 1 , X. Sharon Qi 2

1 Manteia Technologies Co., Ltd, Research Algorithms, Xiamen, China. 2 UCLA, Radiation Oncology, Los Angeles, USA

Purpose/Objective:

Prostate cancer is the most common non-skin cancer for men, with a recurrence rate of approximately 20-50% after primary treatment. Local recurrence, following external beam radiotherapy (EBRT), is most commonly observed in the site of dominant intraprostatic lesion (DIL). Given increasing toxicity with whole prostate stereotactic body radiotherapy (SBRT) doses over 40 Gy, clinical attention has turned to boosting DIL via simultaneous integrated boost (SIB).