Search

Background and Purpose: Auto-contouring of complex anatomy in computed tomography (CT) scans is a highly anticipated solution to many problems in radiotherapy. In this study, artificial intelligence (AI)-based auto-contouring models were clinically validated for lymph node levels and structures of swallowing and chewing in the head and neck. Materials and Methods: CT scans of 145 head and neck radiotherapy patients were retrospectively curated. One cohort (n = 47) was used to analyze seven lymph node levels and the other (n = 98) used to analyze 17 swallowing and chewing structures. Separate nnUnet models were trained and validated using the separate cohorts. For the lymph node levels, preference and clinical acceptability of AI vs human contours were scored. For the swallowing and chewing structures, clinical acceptability was scored. Quantitative analyses of the test sets were performed for AI vs human contours for all structures using overlap and distance metrics. Results: Median Dice Similarity Coefficient ranged from 0.77 to 0.89 for lymph node levels and 0.86 to 0.96 for chewing and swallowing structures. The AI contours were superior to or equally preferred to the manual contours at rates ranging from 75% to 91%; there was not a significant difference in clinical acceptability for nodal levels I-V for manual versus AI contours. Across all AI-generated lymph node level contours, 92% were rated as usable with stylistic to no edits. Of the 340 contours in the chewing and swallowing cohort, 4% required minor edits. Conclusions: An accurate approach was developed to auto-contour lymph node levels and chewing and swallowing structures on CT images for patients with intact nodal anatomy. Only a small portion of test set auto-contours required minor edits.

Background and Purpose: Auto-contouring of complex anatomy in computed tomography (CT) scans is a highly anticipated solution to many problems in radiotherapy. In this study, artificial intelligence (AI)-based auto-contouring models were clinically validated for lymph node levels and structures of swallowing and chewing in the head and neck., Materials and Methods: CT scans of 145 head and neck radiotherapy patients were retrospectively curated. One cohort (n = 47) was used to analyze seven lymph node levels and the other (n = 98) used to analyze 17 swallowing and chewing structures. Separate nnUnet models were trained and validated using the separate cohorts. For the lymph node levels, preference and clinical acceptability of AI vs human contours were scored. For the swallowing and chewing structures, clinical acceptability was scored. Quantitative analyses of the test sets were performed for AI vs human contours for all structures using overlap and distance metrics., Results: Median Dice Similarity Coefficient ranged from 0.77 to 0.89 for lymph node levels and 0.86 to 0.96 for chewing and swallowing structures. The AI contours were superior to or equally preferred to the manual contours at rates ranging from 75% to 91%; there was not a significant difference in clinical acceptability for nodal levels I-V for manual versus AI contours. Across all AI-generated lymph node level contours, 92% were rated as usable with stylistic to no edits. Of the 340 contours in the chewing and swallowing cohort, 4% required minor edits., Conclusions: An accurate approach was developed to auto-contour lymph node levels and chewing and swallowing structures on CT images for patients with intact nodal anatomy. Only a small portion of test set auto-contours required minor edits., Competing Interests: The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Dr. Fuller received/receives related funding and salary support from: National Institutes of Health (NIH) National Cancer Institute (NCI) and  National Institute of Dental and Craniofacial Research (NIDCR) Administrative Supplements to Support Collaborations to Improve AIML-Readiness of NIH-Supported Data (R01CA257814-02S3; R01DE028290-04S2); NIH National Institute of Biomedical Imaging and Bioengineering (NIBIB) Research Education Programs for Residents and Clinical Fellows Grant  (R25EB025787);  NIH/NCI Cancer Center Support Grant (CCSG) (P30CA016672);  Patient-Centered Outcomes Research Institute (PCS-1609–36195;  sub-award from Princess Margaret Hospital). Dr. Fuller receives grant and infrastructure support from MD Anderson Cancer Center via the Charles and Daneen Stiefel Center for Head and Neck Cancer Oropharyngeal Cancer Research Program. Dr. Fuller has received NIH sub-award support from  Oncospace, Inc. (R43CA254559, PI Lakshminarayanan) under a Small Business Innovation Research Grant Applications grant. Dr. Fuller has received direct industry grant/in-kind support, honoraria, and travel funding from Elekta AB. Dr. Fuller has served as a consulting capacity for Varian/Siemens Healthineers, Philips Medical Systems, and Oncospace, Inc.; Dr. Court receives funding (last 36 months) from NCI, CPRIT, Wellcome Trust,The Fund for Innovation in Cancer Informatics, Varian Medical Systems. Dr. Frank has grant funding from Hitach for a Phase II/III Randomized Head and Neck Trial. Dr. Frank is a paid consultant from IBA and Boston Scienfitic. Dr. Frank is the founder and Chair of the Scientific Advisory Committee of C4 Imaging with patent, royalty, and ownership interest. None of the financial interests or personal relationships appear to influence the work reported in this paper. The authors thank the MD Anderson High Performance Computation group and Tumor Measurement Initiative for their support and use of resources., (© 2024 The Author(s).)