Autosegmentation On Low-Resolution T2-Weighted MRI of Head and Neck Cancers for Off-Line Dose Reconstruction in MR-Linac Adapt-To-Position Workflow
B McDonald1*, C Cardenas2, N O'connell3, S Ahmed4, M Naser5, J Xu6, D Thill7, R Zuhour8, S Mesko9, A Augustyn10, S Buszek11, S Grant12, B Chapman13, A Bagley14, K Wahid15, R He16, A Mohamed17, A Dresner18, J Christodouleas19, K Brock20, C Fuller21, (1) UT MD Anderson Cancer Center, Houston, TX, (2) UT MD Anderson Cancer Center, Houston, TX, (3) Elekta Inc., ,,(4) UT MD Anderson Cancer Center, ,,(5) UT MD Anderson Cancer Center, Houston, ,(6) Elekta Inc., St. Charles, MO, (7) Elekta Inc., St. Charles, MO, (8) University Of Texas Medical Branch, ,,(9) UT MD Anderson Cancer Center, ,,(10) UT MD Anderson Cancer Center, ,,(11) UT MD Anderson Cancer Center, ,,(12) UT MD Anderson Cancer Center, ,,(13) UT MD Anderson Cancer Center, ,,(14) UT MD Anderson Cancer Center, ,,(15) UT MD Anderson Cancer Center, ,,(16) UT MD Anderson Cancer Center, Houston, TX, (17) UT MD Anderson Cancer Center, Houston, TX, (18) Philips Healthcare, ,,(19) Elekta Inc., ,,(20) UT MD Anderson Cancer Center, Houston, TX, (21) UT MD Anderson Cancer Center, Houston, TX
Presentations
MO-IePD-TRACK 3-1 (Monday, 7/26/2021) 5:30 PM - 6:00 PM [Eastern Time (GMT-4)]
Purpose: To compare autosegmentation methods on low-resolution T2-weighted on-board setup MRIs from a 1.5T MR-linac for off-line reconstruction of delivered dose.
Methods: 7 organs-at-risk (OARs) (parotids, submandibulars, mandible, cord, brainstem) were contoured by 7 observers each in 43 images. Ground truth (GT) contours were generated using STAPLE. 20 autosegmentation methods were evaluated in ADMIRE: 1-9) atlas-based autosegmentation using a population atlas library (PAL) of 5/10/15 patients with STAPLE, patch fusion (PF), random forest (RF) for label fusion; 10-19) autosegmentation using images from a patient’s 1-4 prior fractions (individualized patient prior (IPP)) using STAPLE/PF/RF; 20) deep learning (DL) (3D ResUNet trained on 43 GT structure sets plus 45 contoured by one observer). Autosegmentation methods were evaluated on 5 images using Dice similarity coefficient, mean surface distance, Hausdorff distance, Jaccard Index against GT. Execution time was measured for each method. Inter-observer variability was quantified using pairwise comparison. For each metric and OAR, performance was compared to inter-observer variability using Dunn’s test with control. Methods were compared pairwise using Steel-Dwass test for each metric pooled across OARs. Further dosimetric analysis was performed on contours from DL (fastest), IPP_RF_4fractions (best performance), and PAL_STAPLE_5atlases (worst performance): Delivered doses from clinical plans were recalculated on setup images with each structure set. Percent differences in Dmean and Dmax were measured between GT and each method.
Results: DL and IPP methods performed best, all significantly outperforming inter-observer variability and with no significant difference between methods in pairwise comparison. Most PAL methods were not significantly different from inter-observer variability or each other. DL was the fastest and PAL methods the slowest. Execution time increased with number of prior fractions/atlases. Dosimetric differences were minimal (median<6%) for DL and IPP_RF_4fractions but greater for PAL_STAPLE_5atlases (median<16%).
Conclusion: Autosegmentation using DL or IPP is superior to PAL considering both geometric and dosimetric criteria.
Funding Support, Disclosures, and Conflict of Interest: This work is directly supported by the NIH/NIDCR (Grant number 1F31DE029093, PI McDonald; grant number 1R01DE028290, PI Fuller) and a Dr. John J. Kopchick Fellowship. Dr. Fuller reports industry research support from Elekta. Ms. McDonald and Dr. Fuller report travel funds and speaking honoraria from Elekta.
ePosters
Keywords
Image-guided Therapy, MRI, Segmentation
Taxonomy
IM/TH- Image Segmentation Techniques: Modality: MRI
Contact Email