C Hague1, A McPartlin2, L W Lee3, C Hughes4, D Mullan5, W Beasley6, A Green7, G Price8, P Whitehurst9, N Slevin10, M van Herk11, C West12, R Chuter13. 1. Department of Head and Neck Clinical Oncology, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Christina.hague@nhs.net. 2. Department of Head and Neck Clinical Oncology, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Andrew.mcpartlin@nhs.net. 3. Department of Head and Neck Clinical Oncology, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: lipwai.lee@nhs.net. 4. Department of Head and Neck Clinical Oncology, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: christopher.hughes@christie.nhs.uk. 5. Department of Radiology, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Damian.mullan@nhs.net. 6. Christie Medical Physics and Engineering, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: w.beasley@nhs.net. 7. Division of Cancer Sciences, Faculty of Biology, Medicine and Heath, University of Manchester, Manchester Academic Health Science Centre, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Andrew.green-2@manchester.ac.uk. 8. Division of Cancer Sciences, Faculty of Biology, Medicine and Heath, University of Manchester, Manchester Academic Health Science Centre, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Gareth.price@manchester.ac.uk. 9. Christie Medical Physics and Engineering, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Philip.whitehurst1@nhs.net. 10. Department of Head and Neck Clinical Oncology, The Christie NHS Foundation Trust, Manchester, UK. 11. Christie Medical Physics and Engineering, The Christie NHS Foundation Trust, Manchester, UK; Division of Cancer Sciences, Faculty of Biology, Medicine and Heath, University of Manchester, Manchester Academic Health Science Centre, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: marcel.vanherk@manchester.ac.uk. 12. Division of Cancer Sciences, Faculty of Biology, Medicine and Heath, University of Manchester, Manchester Academic Health Science Centre, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Catharine.west@manchester.ac.uk. 13. Christie Medical Physics and Engineering, The Christie NHS Foundation Trust, Manchester, UK; Division of Cancer Sciences, Faculty of Biology, Medicine and Heath, University of Manchester, Manchester Academic Health Science Centre, The Christie NHS Foundation Trust, Manchester, UK. Electronic address: Robert.chuter@nhs.net.
Abstract
INTRODUCTION: Auto contouring models help consistently define volumes and reduce clinical workload. This study aimed to evaluate the cross acquisition of a Magnetic Resonance (MR) deep learning auto contouring model for organ at risk (OAR) delineation in head and neck radiotherapy. METHODS: Two auto contouring models were evaluated using deep learning contouring expert (DLCExpert) for OAR delineation: a CT model (modelCT) and an MR model (modelMRI). Models were trained to generate auto contours for the bilateral parotid glands and submandibular glands. Auto-contours for modelMRI were trained on diagnostic images and tested on 10 diagnostic, 10 MR radiotherapy planning (RTP), eight MR-Linac (MRL) scans and, by modelCT, on 10 CT planning scans. Goodness of fit scores, dice similarity coefficient (DSC) and distance to agreement (DTA) were calculated for comparison. RESULTS: ModelMRI contours improved the mean DSC and DTA compared with manual contours for the bilateral parotid glands and submandibular glands on the diagnostic and RTP MRs compared with the MRL sequence. There were statistically significant differences seen for modelMRI compared to modelCT for the left parotid (mean DTA 2.3 v 2.8 mm), right parotid (mean DTA 1.9 v 2.7 mm), left submandibular gland (mean DTA 2.2 v 2.4 mm) and right submandibular gland (mean DTA 1.6 v 3.2 mm). CONCLUSION: A deep learning MR auto-contouring model shows promise for OAR auto-contouring with statistically improved performance vs a CT based model. Performance is affected by the method of MR acquisition and further work is needed to improve its use with MRL images. Crown
INTRODUCTION: Auto contouring models help consistently define volumes and reduce clinical workload. This study aimed to evaluate the cross acquisition of a Magnetic Resonance (MR) deep learning auto contouring model for organ at risk (OAR) delineation in head and neck radiotherapy. METHODS: Two auto contouring models were evaluated using deep learning contouring expert (DLCExpert) for OAR delineation: a CT model (modelCT) and an MR model (modelMRI). Models were trained to generate auto contours for the bilateral parotid glands and submandibular glands. Auto-contours for modelMRI were trained on diagnostic images and tested on 10 diagnostic, 10 MR radiotherapy planning (RTP), eight MR-Linac (MRL) scans and, by modelCT, on 10 CT planning scans. Goodness of fit scores, dice similarity coefficient (DSC) and distance to agreement (DTA) were calculated for comparison. RESULTS: ModelMRI contours improved the mean DSC and DTA compared with manual contours for the bilateral parotid glands and submandibular glands on the diagnostic and RTP MRs compared with the MRL sequence. There were statistically significant differences seen for modelMRI compared to modelCT for the left parotid (mean DTA 2.3 v 2.8 mm), right parotid (mean DTA 1.9 v 2.7 mm), left submandibular gland (mean DTA 2.2 v 2.4 mm) and right submandibular gland (mean DTA 1.6 v 3.2 mm). CONCLUSION: A deep learning MR auto-contouring model shows promise for OAR auto-contouring with statistically improved performance vs a CT based model. Performance is affected by the method of MR acquisition and further work is needed to improve its use with MRL images. Crown