Skip to main content
Top
Published in: European Spine Journal 11/2023

24-04-2023 | Original Article

A deep learning-based technique for the diagnosis of epidural spinal cord compression on thoracolumbar CT

Authors: James Thomas Patrick Decourcy Hallinan, Lei Zhu, Hui Wen Natalie Tan, Si Jian Hui, Xinyi Lim, Bryan Wei Loong Ong, Han Yang Ong, Sterling Ellis Eide, Amanda J. L. Cheng, Shuliang Ge, Tricia Kuah, Shi Wei Desmond Lim, Xi Zhen Low, Ee Chin Teo, Qai Ven Yap, Yiong Huak Chan, Naresh Kumar, Balamurugan A. Vellayappan, Beng Chin Ooi, Swee Tian Quek, Andrew Makmur, Jiong Hao Tan

Published in: European Spine Journal | Issue 11/2023

Login to get access

Abstract

Purpose

To develop a deep learning (DL) model for epidural spinal cord compression (ESCC) on CT, which will aid earlier ESCC diagnosis for less experienced clinicians.

Methods

We retrospectively collected CT and MRI data from adult patients with suspected ESCC at a tertiary referral institute from 2007 till 2020. A total of 183 patients were used for training/validation of the DL model. A separate test set of 40 patients was used for DL model evaluation and comprised 60 staging CT and matched MRI scans performed with an interval of up to 2 months. DL model performance was compared to eight readers: one musculoskeletal radiologist, two body radiologists, one spine surgeon, and four trainee spine surgeons. Diagnostic performance was evaluated using inter-rater agreement, sensitivity, specificity and AUC.

Results

Overall, 3115 axial CT slices were assessed. The DL model showed high kappa of 0.872 for normal, low and high-grade ESCC (trichotomous), which was superior compared to a body radiologist (R4, κ = 0.667) and all four trainee spine surgeons (κ range = 0.625–0.838)(all p < 0.001). In addition, for dichotomous normal versus any grade of ESCC detection, the DL model showed high kappa (κ = 0.879), sensitivity (91.82), specificity (92.01) and AUC (0.919), with the latter AUC superior to all readers (AUC range = 0.732–0.859, all p < 0.001).

Conclusion

A deep learning model for the objective assessment of ESCC on CT had comparable or superior performance to radiologists and spine surgeons. Earlier diagnosis of ESCC on CT could reduce treatment delays, which are associated with poor outcomes, increased costs, and reduced survival.
Appendix
Available only for authorised users
Literature
3.
4.
go back to reference (2008) Metastatic spinal cord compression: diagnosis and management of patients at risk of or with metastatic spinal cord compression. National Collaborating Centre for Cancer, Cardiff (2008) Metastatic spinal cord compression: diagnosis and management of patients at risk of or with metastatic spinal cord compression. National Collaborating Centre for Cancer, Cardiff
5.
go back to reference Hallinan JTPD, Ge S, Zhu L, Zhang W, Lim YT, Thian YL, Jagmohan P, Kuah T, Lim DSW, Low XZ, Teo EC, Barr Kumarakulasinghe N, Yap QV, Chan YH, Tan JH, Kumar N, Vellayappan BA, Ooi BC, Quek ST, Makmur A (2022) Diagnostic accuracy of CT for metastatic epidural spinal cord compression. Cancers (Basel) 14:4231. https://doi.org/10.3390/cancers14174231CrossRefPubMed Hallinan JTPD, Ge S, Zhu L, Zhang W, Lim YT, Thian YL, Jagmohan P, Kuah T, Lim DSW, Low XZ, Teo EC, Barr Kumarakulasinghe N, Yap QV, Chan YH, Tan JH, Kumar N, Vellayappan BA, Ooi BC, Quek ST, Makmur A (2022) Diagnostic accuracy of CT for metastatic epidural spinal cord compression. Cancers (Basel) 14:4231. https://​doi.​org/​10.​3390/​cancers14174231CrossRefPubMed
8.
go back to reference Hallinan JTPD, Zhu L, Yang K, Makmur A, Algazwi DAR, Thian YL, Lau S, Choo YS, Eide SE, Yap QV, Chan YH, Tan JH, Kumar N, Ooi BC, Yoshioka H, Quek ST (2021) Deep learning model for automated detection and classification of central canal, lateral recess, and neural foraminal stenosis at lumbar spine MRI. Radiology 300:130–138. https://doi.org/10.1148/radiol.2021204289CrossRefPubMed Hallinan JTPD, Zhu L, Yang K, Makmur A, Algazwi DAR, Thian YL, Lau S, Choo YS, Eide SE, Yap QV, Chan YH, Tan JH, Kumar N, Ooi BC, Yoshioka H, Quek ST (2021) Deep learning model for automated detection and classification of central canal, lateral recess, and neural foraminal stenosis at lumbar spine MRI. Radiology 300:130–138. https://​doi.​org/​10.​1148/​radiol.​2021204289CrossRefPubMed
9.
go back to reference Lim DSW, Makmur A, Zhu L, Zhang W, Cheng AJL, Sia DSY, Eide SE, Ong HY, Jagmohan P, Tan WC, Khoo VM, Wong YM, Thian YL, Baskar S, Teo EC, Algazwi DAR, Yap QV, Chan YH, Tan JH, Kumar N, Ooi BC, Yoshioka H, Quek ST, Hallinan JTPD (2022) Improved productivity using deep learning-assisted reporting for lumbar spine MRI. Radiology 305:160–166. https://doi.org/10.1148/radiol.220076CrossRefPubMed Lim DSW, Makmur A, Zhu L, Zhang W, Cheng AJL, Sia DSY, Eide SE, Ong HY, Jagmohan P, Tan WC, Khoo VM, Wong YM, Thian YL, Baskar S, Teo EC, Algazwi DAR, Yap QV, Chan YH, Tan JH, Kumar N, Ooi BC, Yoshioka H, Quek ST, Hallinan JTPD (2022) Improved productivity using deep learning-assisted reporting for lumbar spine MRI. Radiology 305:160–166. https://​doi.​org/​10.​1148/​radiol.​220076CrossRefPubMed
13.
go back to reference Hallinan JTPD, Zhu L, Zhang W, Kuah T, Lim DSW, Low XZ, Cheng AJL, Eide SE, Ong HY, Muhamat Nor FE, Alsooreti AM, AlMuhaish MI, Yeong KY, Teo EC, Barr Kumarakulasinghe N, Yap QV, Chan YH, Lin S, Tan JH, Kumar N, Vellayappan BA, Ooi BC, Quek ST, Makmur A (2022) Deep learning model for grading metastatic epidural spinal cord compression on staging CT. Cancers (Basel) 14:3219. https://doi.org/10.3390/cancers14133219CrossRefPubMed Hallinan JTPD, Zhu L, Zhang W, Kuah T, Lim DSW, Low XZ, Cheng AJL, Eide SE, Ong HY, Muhamat Nor FE, Alsooreti AM, AlMuhaish MI, Yeong KY, Teo EC, Barr Kumarakulasinghe N, Yap QV, Chan YH, Lin S, Tan JH, Kumar N, Vellayappan BA, Ooi BC, Quek ST, Makmur A (2022) Deep learning model for grading metastatic epidural spinal cord compression on staging CT. Cancers (Basel) 14:3219. https://​doi.​org/​10.​3390/​cancers14133219CrossRefPubMed
16.
18.
go back to reference Ooi BC, Tan KL, Wang S, Wang W, Cai Q, Chen G, Gao J, Luo Z, Tung AK, Wang Y, Xie Z, Zhang M, Zheng K (2015) SINGA: a distributed deep learning platform. In: Proceedings of the 23rd ACM International Conference on Multimedia, Brisbane, Australia, pp 685–688. https://doi.org/10.1145/2733373.2807410 Ooi BC, Tan KL, Wang S, Wang W, Cai Q, Chen G, Gao J, Luo Z, Tung AK, Wang Y, Xie Z, Zhang M, Zheng K (2015) SINGA: a distributed deep learning platform. In: Proceedings of the 23rd ACM International Conference on Multimedia, Brisbane, Australia, pp 685–688. https://​doi.​org/​10.​1145/​2733373.​2807410
19.
go back to reference Luo Z, Yeung SH, Zhang M, Zheng K, Zhu L, Chen G, Fan F, Lin Q, Ngiam KY, Ooi BC (2021) MLCask: efficient management of component evolution in collaborative data analytics pipelines. In IEEE 37th International Conference on Data Engineering (ICDE). Chania, Crete, Greece, pp 1655–1666. https://doi.org/10.1109/ICDE51399.2021.00146 Luo Z, Yeung SH, Zhang M, Zheng K, Zhu L, Chen G, Fan F, Lin Q, Ngiam KY, Ooi BC (2021) MLCask: efficient management of component evolution in collaborative data analytics pipelines. In IEEE 37th International Conference on Data Engineering (ICDE). Chania, Crete, Greece, pp 1655–1666. https://​doi.​org/​10.​1109/​ICDE51399.​2021.​00146
25.
go back to reference Glicksman RM, Tjong MC, Neves-Junior WFP, Spratt DE, Chua KLM, Mansouri A, Chua MLK, Berlin A, Winter JD, Dahele M, Slotman BJ, Bilsky M, Shultz DB, Maldaun M, Szerlip N, Lo SS, Yamada Y, Vera-Badillo FE, Marta GN, Moraes FY (2020) Stereotactic ablative radiotherapy for the management of spinal metastases: a review. JAMA Oncol 6:567–577. https://doi.org/10.1001/jamaoncol.2019.5351CrossRefPubMed Glicksman RM, Tjong MC, Neves-Junior WFP, Spratt DE, Chua KLM, Mansouri A, Chua MLK, Berlin A, Winter JD, Dahele M, Slotman BJ, Bilsky M, Shultz DB, Maldaun M, Szerlip N, Lo SS, Yamada Y, Vera-Badillo FE, Marta GN, Moraes FY (2020) Stereotactic ablative radiotherapy for the management of spinal metastases: a review. JAMA Oncol 6:567–577. https://​doi.​org/​10.​1001/​jamaoncol.​2019.​5351CrossRefPubMed
Metadata
Title
A deep learning-based technique for the diagnosis of epidural spinal cord compression on thoracolumbar CT
Authors
James Thomas Patrick Decourcy Hallinan
Lei Zhu
Hui Wen Natalie Tan
Si Jian Hui
Xinyi Lim
Bryan Wei Loong Ong
Han Yang Ong
Sterling Ellis Eide
Amanda J. L. Cheng
Shuliang Ge
Tricia Kuah
Shi Wei Desmond Lim
Xi Zhen Low
Ee Chin Teo
Qai Ven Yap
Yiong Huak Chan
Naresh Kumar
Balamurugan A. Vellayappan
Beng Chin Ooi
Swee Tian Quek
Andrew Makmur
Jiong Hao Tan
Publication date
24-04-2023
Publisher
Springer Berlin Heidelberg
Published in
European Spine Journal / Issue 11/2023
Print ISSN: 0940-6719
Electronic ISSN: 1432-0932
DOI
https://doi.org/10.1007/s00586-023-07706-4

Other articles of this Issue 11/2023

European Spine Journal 11/2023 Go to the issue