Jordan Wong1, Allan Fong2, Nevin McVicar3, Sally Smith4, Joshua Giambattista5, Derek Wells6, Carter Kolbeck7, Jonathan Giambattista8, Lovedeep Gondara9, Abraham Alexander10. 1. BC Cancer - Vancouver Center, Canada. Electronic address: Jordan.wong@bccancer.bc.ca. 2. BC Cancer - Vancouver Center, Canada. Electronic address: allan.fong1@bccancer.bc.ca. 3. BC Cancer - Vancouver Center, Canada. Electronic address: mcvicarn@rvh.on.ca. 4. BC Cancer - Victoria Center, Canada. Electronic address: ssmith11@bccancer.bc.ca. 5. Saskatchewan Cancer Agency, Regina, Canada; Limbus AI Inc., Regina, Canada. Electronic address: joshua.giambattista@saskcancer.ca. 6. BC Cancer - Victoria Center, Canada. Electronic address: DWells@bccancer.bc.ca. 7. Limbus AI Inc., Regina, Canada. Electronic address: carter@limbus.ai. 8. Limbus AI Inc., Regina, Canada. Electronic address: jon@limbus.ai. 9. BC Cancer - Vancouver Center, Canada. Electronic address: Lovedeep.Gondara@bccancer.bc.ca. 10. BC Cancer - Victoria Center, Canada. Electronic address: AAlexander3@bccancer.bc.ca.
Abstract
BACKGROUND: Deep learning-based auto-segmented contours (DC) aim to alleviate labour intensive contouring of organs at risk (OAR) and clinical target volumes (CTV). Most previous DC validation studies have a limited number of expert observers for comparison and/or use a validation dataset related to the training dataset. We determine if DC models are comparable to Radiation Oncologist (RO) inter-observer variability on an independent dataset. METHODS: Expert contours (EC) were created by multiple ROs for central nervous system (CNS), head and neck (H&N), and prostate radiotherapy (RT) OARs and CTVs. DCs were generated using deep learning-based auto-segmentation software trained by a single RO on publicly available data. Contours were compared using Dice Similarity Coefficient (DSC) and 95% Hausdorff distance (HD). RESULTS: Sixty planning CT scans had 2-4 ECs, for a total of 60 CNS, 53 H&N, and 50 prostate RT contour sets. The mean DC and EC contouring times were 0.4 vs 7.7 min for CNS, 0.6 vs 26.6 min for H&N, and 0.4 vs 21.3 min for prostate RT contours. There were minimal differences in DSC and 95% HD involving DCs for OAR comparisons, but more noticeable differences for CTV comparisons. CONCLUSIONS: The accuracy of DCs trained by a single RO is comparable to expert inter-observer variability for the RT planning contours in this study. Use of deep learning-based auto-segmentation in clinical practice will likely lead to significant benefits to RT planning workflow and resources.
BACKGROUND: Deep learning-based auto-segmented contours (DC) aim to alleviate labour intensive contouring of organs at risk (OAR) and clinical target volumes (CTV). Most previous DC validation studies have a limited number of expert observers for comparison and/or use a validation dataset related to the training dataset. We determine if DC models are comparable to Radiation Oncologist (RO) inter-observer variability on an independent dataset. METHODS: Expert contours (EC) were created by multiple ROs for central nervous system (CNS), head and neck (H&N), and prostate radiotherapy (RT) OARs and CTVs. DCs were generated using deep learning-based auto-segmentation software trained by a single RO on publicly available data. Contours were compared using Dice Similarity Coefficient (DSC) and 95% Hausdorff distance (HD). RESULTS: Sixty planning CT scans had 2-4 ECs, for a total of 60 CNS, 53 H&N, and 50 prostate RT contour sets. The mean DC and EC contouring times were 0.4 vs 7.7 min for CNS, 0.6 vs 26.6 min for H&N, and 0.4 vs 21.3 min for prostate RT contours. There were minimal differences in DSC and 95% HD involving DCs for OAR comparisons, but more noticeable differences for CTV comparisons. CONCLUSIONS: The accuracy of DCs trained by a single RO is comparable to expert inter-observer variability for the RT planning contours in this study. Use of deep learning-based auto-segmentation in clinical practice will likely lead to significant benefits to RT planning workflow and resources.
Authors: Asma Amjad; Jiaofeng Xu; Dan Thill; Colleen Lawton; William Hall; Musaddiq J Awan; Monica Shukla; Beth A Erickson; X Allen Li Journal: Med Phys Date: 2022-02-07 Impact factor: 4.071
Authors: Jeremy M Webb; Shaheeda A Adusei; Yinong Wang; Naziya Samreen; Kalie Adler; Duane D Meixner; Robert T Fazzio; Mostafa Fatemi; Azra Alizad Journal: Comput Biol Med Date: 2021-10-21 Impact factor: 4.589
Authors: Ana Barragán-Montero; Umair Javaid; Gilmer Valdés; Dan Nguyen; Paul Desbordes; Benoit Macq; Siri Willems; Liesbeth Vandewinckele; Mats Holmström; Fredrik Löfman; Steven Michiels; Kevin Souris; Edmond Sterpin; John A Lee Journal: Phys Med Date: 2021-05-09 Impact factor: 2.685
Authors: Elaine Cha; Sharif Elguindi; Ifeanyirochukwu Onochie; Daniel Gorovets; Joseph O Deasy; Michael Zelefsky; Erin F Gillespie Journal: Radiother Oncol Date: 2021-03-03 Impact factor: 6.901
Authors: Danju Huang; Han Bai; Li Wang; Yu Hou; Lan Li; Yaoxiong Xia; Zhirui Yan; Wenrui Chen; Li Chang; Wenhui Li Journal: Technol Cancer Res Treat Date: 2021 Jan-Dec