Generating High-Quality Lymph Node Clinical Target Volumes for Head and Neck Cancer Radiotherapy Using a Fully Automated Deep Learning-Based Approach. International journal of radiation oncology, biology, physics Cardenas, C. E., Beadle, B. M., Garden, A. S., Skinner, H. D., Yang, J., Rhee, D. J., McCarroll, R. E., Netherton, T. J., Gay, S. S., Zhang, L., Court, L. E. 2020


PURPOSE: To develop a deep learning model that generates consistent, high-quality lymph node CTV contours for HNC patients, as an integral part of a fully-automated radiation treatment planning workflow.METHODS AND MATERIALS: CT scans from 71 HNC patients were retrospectively collected and split into training (n=51), cross-validation (n=10), and test (n=10) datasets. All had target volume delineations covering lymph node levels Ia through V (Ia-V), Ib through V (Ib-V), II through IV (II-IV), and retropharyngeal (RP) nodes, which were previously approved by a radiation oncologist specializing in HNC. Volumes of interest (VOIs) about nodal levels were automatically identified using computer vision techniques. The VOI (cropped CT image) and approved contours were used to train a U-Net auto-segmentation model. Each lymph node level was trained independently, with model parameters optimized by assessing performance on the cross-validation dataset. Once optimal model parameters were identified, overlap and distance metrics were calculated between ground truth and auto-segmentations on the test set. Lastly, this final model was used on 32 additional patient scans (not included in original 71 cases) and auto-segmentations visually rated by three radiation oncologists as being "clinically acceptable without requiring edits", "requiring minor edits", or "requiring major edits."RESULTS: When comparing ground truths to auto-segmentations on the test dataset, median Dice Similarity Coefficients were 0.90, 0.90, 0.89, and 0.81 and median mean surface distance values were 1.0mm, 1.0mm, 1.1mm, and 1.3mm for node levels Ia-V, Ib-V, II-IV, and RP nodes, respectively. Qualitative scoring varied between physicians. Overall, 99% of auto-segmented target volumes were either scored as being clinically acceptable or requiring minor edits (i.e. stylistic recommendations, <2 minutes).CONCLUSIONS: We developed a fully automated artificial intelligence approach to auto-delineate nodal CTVs for patients with intact HNC. Most auto-segmentations were found to be clinically acceptable after qualitative review when considering recommended stylistic edits. This promising work automatically delineates nodal CTVs in a robust and consistent manner; this approach can be implemented in ongoing efforts for fully-automated radiation treatment planning.

View details for DOI 10.1016/j.ijrobp.2020.10.005

View details for PubMedID 33068690