Evan Porter1,2,3, Patricia Fuentes2,4, Zaid Siddiqui2,3, Andrew Thompson2,3, Ronald Levitin2,3, David Solis2,3, Nick Myziuk2,3, Thomas Guerrero2,3,4. 1. Department of Medical Physics, Wayne State University, Detroit, MI, USA. 2. Beaumont Artificial Intelligence Research Laboratory, Beaumont Health Systems, Royal Oak, MI, USA. 3. Department of Radiation Oncology, Beaumont Health Systems, Royal Oak, MI, USA. 4. Oakland University William Beaumont School of Medicine, Oakland University, Rochester, MI, USA.
Abstract
PURPOSE: Accurate segmentation of the hippocampus for hippocampal avoidance whole-brain radiotherapy currently requires high-resolution magnetic resonance imaging (MRI) in addition to neuroanatomic expertise for manual segmentation. Removing the need for MR images to identify the hippocampus would reduce planning complexity, the need for a treatment planning MR imaging session, potential uncertainties associated with MRI-computed tomography (CT) image registration, and cost. Three-dimensional (3D) deep convolutional network models have the potential to automate hippocampal segmentation. In this study, we investigate the accuracy and reliability of hippocampal segmentation by automated deep learning models from CT alone and compare the accuracy to experts using MRI fusion. METHODS: Retrospectively, 390 Gamma Knife patients with high-resolution CT and MR images were collected. Following the RTOG 0933 guidelines, images were rigidly fused, and a neuroanatomic expert contoured the hippocampus on the MR, then transferred the contours to CT. Using a calculated cranial centroid, the image volumes were cropped to 200 × 200 × 35 voxels, which were used to train four models, including our proposed Attention-Gated 3D ResNet (AG-3D ResNet). These models were then compared with results from a nested tenfold validation. From the predicted test set volumes, we calculated the 100% Hausdorff distance (HD). Acceptability was assessed using the RTOG 0933 protocol criteria, and contours were considered passing with HD ≤ 7 mm. RESULTS: The bilateral hippocampus passing rate across all 90 models trained in the nested cross-fold validation was 80.2% for AG-3D ResNet, which performs with a comparable pass rate (P = 0.3345) to physicians during centralized review for the RTOG 0933 Phase II clinical trial. CONCLUSIONS: Our proposed AG-3D ResNet's segmentation of the hippocampus from noncontrast CT images alone are comparable to those obtained by participating physicians from the RTOG 0933 Phase II clinical trial.
PURPOSE: Accurate segmentation of the hippocampus for hippocampal avoidance whole-brain radiotherapy currently requires high-resolution magnetic resonance imaging (MRI) in addition to neuroanatomic expertise for manual segmentation. Removing the need for MR images to identify the hippocampus would reduce planning complexity, the need for a treatment planning MR imaging session, potential uncertainties associated with MRI-computed tomography (CT) image registration, and cost. Three-dimensional (3D) deep convolutional network models have the potential to automate hippocampal segmentation. In this study, we investigate the accuracy and reliability of hippocampal segmentation by automated deep learning models from CT alone and compare the accuracy to experts using MRI fusion. METHODS: Retrospectively, 390 Gamma Knife patients with high-resolution CT and MR images were collected. Following the RTOG 0933 guidelines, images were rigidly fused, and a neuroanatomic expert contoured the hippocampus on the MR, then transferred the contours to CT. Using a calculated cranial centroid, the image volumes were cropped to 200 × 200 × 35 voxels, which were used to train four models, including our proposed Attention-Gated 3D ResNet (AG-3D ResNet). These models were then compared with results from a nested tenfold validation. From the predicted test set volumes, we calculated the 100% Hausdorff distance (HD). Acceptability was assessed using the RTOG 0933 protocol criteria, and contours were considered passing with HD ≤ 7 mm. RESULTS: The bilateral hippocampus passing rate across all 90 models trained in the nested cross-fold validation was 80.2% for AG-3D ResNet, which performs with a comparable pass rate (P = 0.3345) to physicians during centralized review for the RTOG 0933 Phase II clinical trial. CONCLUSIONS: Our proposed AG-3D ResNet's segmentation of the hippocampus from noncontrast CT images alone are comparable to those obtained by participating physicians from the RTOG 0933 Phase II clinical trial.
Authors: Annika Hänsch; Jan Hendrik Moltz; Benjamin Geisler; Christiane Engel; Jan Klein; Angelo Genghi; Jan Schreier; Tomasz Morgas; Benjamin Haas Journal: J Med Imaging (Bellingham) Date: 2020-11-11