Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI
BackgroundMetastatic epidural spinal cord compression (MESCC) is a devastating complication of advanced cancer. A deep learning (DL) model for automated MESCC classification on MRI could aid earlier diagnosis and referral.PurposeTo develop a DL model for automated classification of MESCC on MRI.Mate...
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-05-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2022.849447/full |
| _version_ | 1818275422460182528 |
|---|---|
| author | James Thomas Patrick Decourcy Hallinan James Thomas Patrick Decourcy Hallinan Lei Zhu Wenqiao Zhang Desmond Shi Wei Lim Desmond Shi Wei Lim Sangeetha Baskar Xi Zhen Low Xi Zhen Low Kuan Yuen Yeong Ee Chin Teo Nesaretnam Barr Kumarakulasinghe Qai Ven Yap Yiong Huak Chan Shuxun Lin Jiong Hao Tan Naresh Kumar Balamurugan A. Vellayappan Beng Chin Ooi Swee Tian Quek Swee Tian Quek Andrew Makmur Andrew Makmur |
| author_facet | James Thomas Patrick Decourcy Hallinan James Thomas Patrick Decourcy Hallinan Lei Zhu Wenqiao Zhang Desmond Shi Wei Lim Desmond Shi Wei Lim Sangeetha Baskar Xi Zhen Low Xi Zhen Low Kuan Yuen Yeong Ee Chin Teo Nesaretnam Barr Kumarakulasinghe Qai Ven Yap Yiong Huak Chan Shuxun Lin Jiong Hao Tan Naresh Kumar Balamurugan A. Vellayappan Beng Chin Ooi Swee Tian Quek Swee Tian Quek Andrew Makmur Andrew Makmur |
| author_sort | James Thomas Patrick Decourcy Hallinan |
| collection | DOAJ |
| description | BackgroundMetastatic epidural spinal cord compression (MESCC) is a devastating complication of advanced cancer. A deep learning (DL) model for automated MESCC classification on MRI could aid earlier diagnosis and referral.PurposeTo develop a DL model for automated classification of MESCC on MRI.Materials and MethodsPatients with known MESCC diagnosed on MRI between September 2007 and September 2017 were eligible. MRI studies with instrumentation, suboptimal image quality, and non-thoracic regions were excluded. Axial T2-weighted images were utilized. The internal dataset split was 82% and 18% for training/validation and test sets, respectively. External testing was also performed. Internal training/validation data were labeled using the Bilsky MESCC classification by a musculoskeletal radiologist (10-year experience) and a neuroradiologist (5-year experience). These labels were used to train a DL model utilizing a prototypical convolutional neural network. Internal and external test sets were labeled by the musculoskeletal radiologist as the reference standard. For assessment of DL model performance and interobserver variability, test sets were labeled independently by the neuroradiologist (5-year experience), a spine surgeon (5-year experience), and a radiation oncologist (11-year experience). Inter-rater agreement (Gwet’s kappa) and sensitivity/specificity were calculated.ResultsOverall, 215 MRI spine studies were analyzed [164 patients, mean age = 62 ± 12(SD)] with 177 (82%) for training/validation and 38 (18%) for internal testing. For internal testing, the DL model and specialists all showed almost perfect agreement (kappas = 0.92–0.98, p < 0.001) for dichotomous Bilsky classification (low versus high grade) compared to the reference standard. Similar performance was seen for external testing on a set of 32 MRI spines with the DL model and specialists all showing almost perfect agreement (kappas = 0.94–0.95, p < 0.001) compared to the reference standard.ConclusionA DL model showed comparable agreement to a subspecialist radiologist and clinical specialists for the classification of malignant epidural spinal cord compression and could optimize earlier diagnosis and surgical referral. |
| first_indexed | 2024-12-12T22:29:30Z |
| format | Article |
| id | doaj.art-5adf8341c6024984a153838ea1686524 |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-12-12T22:29:30Z |
| publishDate | 2022-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-5adf8341c6024984a153838ea16865242022-12-22T00:09:39ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2022-05-011210.3389/fonc.2022.849447849447Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRIJames Thomas Patrick Decourcy Hallinan0James Thomas Patrick Decourcy Hallinan1Lei Zhu2Wenqiao Zhang3Desmond Shi Wei Lim4Desmond Shi Wei Lim5Sangeetha Baskar6Xi Zhen Low7Xi Zhen Low8Kuan Yuen Yeong9Ee Chin Teo10Nesaretnam Barr Kumarakulasinghe11Qai Ven Yap12Yiong Huak Chan13Shuxun Lin14Jiong Hao Tan15Naresh Kumar16Balamurugan A. Vellayappan17Beng Chin Ooi18Swee Tian Quek19Swee Tian Quek20Andrew Makmur21Andrew Makmur22Department of Diagnostic Imaging, National University Hospital, Singapore, SingaporeDepartment of Diagnostic Radiology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, SingaporeNUS Graduate School, Integrative Sciences and Engineering Programme, National University of Singapore, Singapore, SingaporeDepartment of Computer Science, School of Computing, National University of Singapore, Singapore, SingaporeDepartment of Diagnostic Imaging, National University Hospital, Singapore, SingaporeDepartment of Diagnostic Radiology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, SingaporeDepartment of Diagnostic Imaging, National University Hospital, Singapore, SingaporeDepartment of Diagnostic Imaging, National University Hospital, Singapore, SingaporeDepartment of Diagnostic Radiology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, SingaporeDepartment of Radiology, Ng Teng Fong General Hospital, Singapore, SingaporeDepartment of Diagnostic Imaging, National University Hospital, Singapore, SingaporeNational University Cancer Institute, NUH Medical Centre (NUHMC), Singapore, SingaporeBiostatistics Unit, Yong Loo Lin School of Medicine, Singapore, SingaporeBiostatistics Unit, Yong Loo Lin School of Medicine, Singapore, SingaporeDivision of Spine Surgery, Department of Orthopaedic Surgery, Ng Teng Fong General Hospital, Singapore, SingaporeUniversity Spine Centre, Department of Orthopaedic Surgery, National University Health System, Singapore, SingaporeUniversity Spine Centre, Department of Orthopaedic Surgery, National University Health System, Singapore, Singapore0Department of Radiation Oncology, National University Cancer Institute Singapore, National University Hospital, Singapore, SingaporeDepartment of Computer Science, School of Computing, National University of Singapore, Singapore, SingaporeDepartment of Diagnostic Imaging, National University Hospital, Singapore, SingaporeDepartment of Diagnostic Radiology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, SingaporeDepartment of Diagnostic Imaging, National University Hospital, Singapore, SingaporeDepartment of Diagnostic Radiology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, SingaporeBackgroundMetastatic epidural spinal cord compression (MESCC) is a devastating complication of advanced cancer. A deep learning (DL) model for automated MESCC classification on MRI could aid earlier diagnosis and referral.PurposeTo develop a DL model for automated classification of MESCC on MRI.Materials and MethodsPatients with known MESCC diagnosed on MRI between September 2007 and September 2017 were eligible. MRI studies with instrumentation, suboptimal image quality, and non-thoracic regions were excluded. Axial T2-weighted images were utilized. The internal dataset split was 82% and 18% for training/validation and test sets, respectively. External testing was also performed. Internal training/validation data were labeled using the Bilsky MESCC classification by a musculoskeletal radiologist (10-year experience) and a neuroradiologist (5-year experience). These labels were used to train a DL model utilizing a prototypical convolutional neural network. Internal and external test sets were labeled by the musculoskeletal radiologist as the reference standard. For assessment of DL model performance and interobserver variability, test sets were labeled independently by the neuroradiologist (5-year experience), a spine surgeon (5-year experience), and a radiation oncologist (11-year experience). Inter-rater agreement (Gwet’s kappa) and sensitivity/specificity were calculated.ResultsOverall, 215 MRI spine studies were analyzed [164 patients, mean age = 62 ± 12(SD)] with 177 (82%) for training/validation and 38 (18%) for internal testing. For internal testing, the DL model and specialists all showed almost perfect agreement (kappas = 0.92–0.98, p < 0.001) for dichotomous Bilsky classification (low versus high grade) compared to the reference standard. Similar performance was seen for external testing on a set of 32 MRI spines with the DL model and specialists all showing almost perfect agreement (kappas = 0.94–0.95, p < 0.001) compared to the reference standard.ConclusionA DL model showed comparable agreement to a subspecialist radiologist and clinical specialists for the classification of malignant epidural spinal cord compression and could optimize earlier diagnosis and surgical referral.https://www.frontiersin.org/articles/10.3389/fonc.2022.849447/fulldeep learning modelmetastatic epidural spinal cord compressionMRIBilsky classificationspinal metastasis classificationspinal metastatic disease |
| spellingShingle | James Thomas Patrick Decourcy Hallinan James Thomas Patrick Decourcy Hallinan Lei Zhu Wenqiao Zhang Desmond Shi Wei Lim Desmond Shi Wei Lim Sangeetha Baskar Xi Zhen Low Xi Zhen Low Kuan Yuen Yeong Ee Chin Teo Nesaretnam Barr Kumarakulasinghe Qai Ven Yap Yiong Huak Chan Shuxun Lin Jiong Hao Tan Naresh Kumar Balamurugan A. Vellayappan Beng Chin Ooi Swee Tian Quek Swee Tian Quek Andrew Makmur Andrew Makmur Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI Frontiers in Oncology deep learning model metastatic epidural spinal cord compression MRI Bilsky classification spinal metastasis classification spinal metastatic disease |
| title | Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI |
| title_full | Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI |
| title_fullStr | Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI |
| title_full_unstemmed | Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI |
| title_short | Deep Learning Model for Classifying Metastatic Epidural Spinal Cord Compression on MRI |
| title_sort | deep learning model for classifying metastatic epidural spinal cord compression on mri |
| topic | deep learning model metastatic epidural spinal cord compression MRI Bilsky classification spinal metastasis classification spinal metastatic disease |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2022.849447/full |
| work_keys_str_mv | AT jamesthomaspatrickdecourcyhallinan deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT jamesthomaspatrickdecourcyhallinan deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT leizhu deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT wenqiaozhang deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT desmondshiweilim deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT desmondshiweilim deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT sangeethabaskar deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT xizhenlow deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT xizhenlow deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT kuanyuenyeong deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT eechinteo deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT nesaretnambarrkumarakulasinghe deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT qaivenyap deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT yionghuakchan deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT shuxunlin deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT jionghaotan deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT nareshkumar deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT balamuruganavellayappan deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT bengchinooi deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT sweetianquek deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT sweetianquek deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT andrewmakmur deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri AT andrewmakmur deeplearningmodelforclassifyingmetastaticepiduralspinalcordcompressiononmri |