Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy

Various commercial auto-contouring solutions have emerged over past few years to address labor-intensiveness, and inter- and intra-operator variabilities issues of traditional manual anatomy contouring for head and neck (H&N) radiation therapy (RT). The purpose of this study is to compare the cl...

Full description

Bibliographic Details
Main Authors: Ng, Curtise, Leung, Vincent WS, Hung, Rico HM
Other Authors: Huang, Yujie
Format: Journal Article
Published: MDPI AG 2022
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/89669
_version_ 1848765267620921344
author Ng, Curtise
Leung, Vincent WS
Hung, Rico HM
author2 Huang, Yujie
author_facet Huang, Yujie
Ng, Curtise
Leung, Vincent WS
Hung, Rico HM
author_sort Ng, Curtise
building Curtin Institutional Repository
collection Online Access
description Various commercial auto-contouring solutions have emerged over past few years to address labor-intensiveness, and inter- and intra-operator variabilities issues of traditional manual anatomy contouring for head and neck (H&N) radiation therapy (RT). The purpose of this study is to compare the clinical performances between RaySearch Laboratories deep learning (DL) and atlas-based auto-contouring tools for organs at risk (OARs) segmentation in the H&N RT with the manual contouring as reference. Forty-five H&N computed tomography datasets were used for the DL and atlas-based auto-contouring tools to contour 16 OARs and time required for the segmentation was measured. Dice similarity coefficient (DSC), Hausdorff distance (HD) and HD 95th-percentile (HD95) were used to evaluate geometric accuracy of OARs contoured by the DL and atlas-based auto-contouring tools. Paired sample t-test was employed to compare the mean DSC, HD, HD95, and contouring time values of the two groups. The DL auto-contouring approach achieved more consistent performance in OARs segmentation than its atlas-based approach, resulting in statistically significant time reduction of the whole segmentation process by 40% (p < 0.001). The DL auto-contouring had statistically significantly higher mean DSC and lower HD and HD95 values (p < 0.001–0.009) for 10 out of 16 OARs. This study proves that the RaySearch Laboratories DL auto-contouring tool has significantly better clinical performances than its atlas-based approach.
first_indexed 2025-11-14T11:32:32Z
format Journal Article
id curtin-20.500.11937-89669
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T11:32:32Z
publishDate 2022
publisher MDPI AG
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-896692023-01-23T05:43:03Z Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy Ng, Curtise Leung, Vincent WS Hung, Rico HM Huang, Yujie Artificial Intelligence Automation Computed Tomography Image Segmentation Intensity-Modulated Radiation Therapy Machine Learning Nasopharyngeal Cancer Organs at Risk Radiotherapy Volumetric Arc Therapy Various commercial auto-contouring solutions have emerged over past few years to address labor-intensiveness, and inter- and intra-operator variabilities issues of traditional manual anatomy contouring for head and neck (H&N) radiation therapy (RT). The purpose of this study is to compare the clinical performances between RaySearch Laboratories deep learning (DL) and atlas-based auto-contouring tools for organs at risk (OARs) segmentation in the H&N RT with the manual contouring as reference. Forty-five H&N computed tomography datasets were used for the DL and atlas-based auto-contouring tools to contour 16 OARs and time required for the segmentation was measured. Dice similarity coefficient (DSC), Hausdorff distance (HD) and HD 95th-percentile (HD95) were used to evaluate geometric accuracy of OARs contoured by the DL and atlas-based auto-contouring tools. Paired sample t-test was employed to compare the mean DSC, HD, HD95, and contouring time values of the two groups. The DL auto-contouring approach achieved more consistent performance in OARs segmentation than its atlas-based approach, resulting in statistically significant time reduction of the whole segmentation process by 40% (p < 0.001). The DL auto-contouring had statistically significantly higher mean DSC and lower HD and HD95 values (p < 0.001–0.009) for 10 out of 16 OARs. This study proves that the RaySearch Laboratories DL auto-contouring tool has significantly better clinical performances than its atlas-based approach. 2022 Journal Article http://hdl.handle.net/20.500.11937/89669 10.3390/app122211681 http://creativecommons.org/licenses/by/4.0/ MDPI AG fulltext
spellingShingle Artificial Intelligence
Automation
Computed Tomography
Image Segmentation
Intensity-Modulated Radiation Therapy
Machine Learning
Nasopharyngeal Cancer
Organs at Risk
Radiotherapy
Volumetric Arc Therapy
Ng, Curtise
Leung, Vincent WS
Hung, Rico HM
Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy
title Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy
title_full Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy
title_fullStr Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy
title_full_unstemmed Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy
title_short Clinical Evaluation of Deep Learning and Atlas-Based Auto-Contouring for Head and Neck Radiation Therapy
title_sort clinical evaluation of deep learning and atlas-based auto-contouring for head and neck radiation therapy
topic Artificial Intelligence
Automation
Computed Tomography
Image Segmentation
Intensity-Modulated Radiation Therapy
Machine Learning
Nasopharyngeal Cancer
Organs at Risk
Radiotherapy
Volumetric Arc Therapy
url http://hdl.handle.net/20.500.11937/89669