Deep Learning Algorithm for Identifying Cervical Cord Compression Due to Degenerative Canal Stenosis on Radiography.
Journal
Spine
ISSN: 1528-1159
Titre abrégé: Spine (Phila Pa 1976)
Pays: United States
ID NLM: 7610646
Informations de publication
Date de publication:
15 Apr 2023
15 Apr 2023
Historique:
received:
21
11
2022
accepted:
06
01
2023
medline:
30
3
2023
pubmed:
11
2
2023
entrez:
10
2
2023
Statut:
ppublish
Résumé
Cross-sectional study. Validate the diagnostic accuracy of a deep-learning algorithm for cervical cord compression due to degenerative canal stenosis on radiography. The diagnosis of degenerative cervical myelopathy is often delayed, resulting in improper management. Screening tools for suspected degenerative cervical myelopathy would help identify patients who require detailed physical evaluation. Data from 240 patients (120 with cervical stenosis on magnetic resonance imaging and 120 age and sex-matched controls) were randomly divided into training (n = 198) and test (n = 42) data sets. The deep-learning algorithm, designed to identify the suspected stenosis level on radiography, was constructed using a convolutional neural network model called EfficientNetB2, and radiography and magnetic resonance imaging data from the training data set. The accuracy and area under the curve of the receiver operating characteristic curve were calculated for the independent test data set. Finally, the number of correct diagnoses was compared between the algorithm and 10 physicians using the test cohort. The diagnostic accuracy and area under the curve of the deep-learning algorithm were 0.81 and 0.81, respectively, in the independent test data set. The rate of correct responses in the test data set was significantly higher for the algorithm than for the physician's consensus (81.0% vs . 66.2%; P = 0.034). Furthermore, the accuracy of the algorithm was greater than that of each individual physician. We developed a deep-learning algorithm capable of suggesting the presence of cervical spinal cord compression on cervical radiography and highlighting the suspected levels on radiographic imaging when cord compression is identified. The diagnostic accuracy of the algorithm was greater than that of spine physicians. Level IV.
Sections du résumé
STUDY DESIGN
METHODS
Cross-sectional study.
OBJECTIVE
OBJECTIVE
Validate the diagnostic accuracy of a deep-learning algorithm for cervical cord compression due to degenerative canal stenosis on radiography.
SUMMARY OF BACKGROUND DATA
BACKGROUND
The diagnosis of degenerative cervical myelopathy is often delayed, resulting in improper management. Screening tools for suspected degenerative cervical myelopathy would help identify patients who require detailed physical evaluation.
MATERIALS AND METHODS
METHODS
Data from 240 patients (120 with cervical stenosis on magnetic resonance imaging and 120 age and sex-matched controls) were randomly divided into training (n = 198) and test (n = 42) data sets. The deep-learning algorithm, designed to identify the suspected stenosis level on radiography, was constructed using a convolutional neural network model called EfficientNetB2, and radiography and magnetic resonance imaging data from the training data set. The accuracy and area under the curve of the receiver operating characteristic curve were calculated for the independent test data set. Finally, the number of correct diagnoses was compared between the algorithm and 10 physicians using the test cohort.
RESULTS
RESULTS
The diagnostic accuracy and area under the curve of the deep-learning algorithm were 0.81 and 0.81, respectively, in the independent test data set. The rate of correct responses in the test data set was significantly higher for the algorithm than for the physician's consensus (81.0% vs . 66.2%; P = 0.034). Furthermore, the accuracy of the algorithm was greater than that of each individual physician.
CONCLUSIONS
CONCLUSIONS
We developed a deep-learning algorithm capable of suggesting the presence of cervical spinal cord compression on cervical radiography and highlighting the suspected levels on radiographic imaging when cord compression is identified. The diagnostic accuracy of the algorithm was greater than that of spine physicians.
LEVEL OF EVIDENCE
METHODS
Level IV.
Identifiants
pubmed: 36763843
doi: 10.1097/BRS.0000000000004595
pii: 00007632-202304150-00002
doi:
Types de publication
Randomized Controlled Trial
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
519-525Informations de copyright
Copyright © 2023 Wolters Kluwer Health, Inc. All rights reserved.
Déclaration de conflit d'intérêts
The authors report no conflicts of interest.
Références
Nouri A, Tetreault L, Singh A, et al. Degenerative cervical myelopathy: epidemiology, genetics, and pathogenesis. Spine (Phila Pa 1976). 2015;40:E675–93.
Witiw CD, Fehlings MG. Degenerative cervical myelopathy. CMAJ. 2017;189:E116.
Tamai K, Suzuki A, Terai H, et al. Change in physical and mental quality-of-life between the short- and mid-term periods after cervical laminoplasty for cervical spondylotic myelopathy: a retrospective cohort study with minimum 5 years follow-up. J Clin Med. 2022;11:5230.
Tamai K, Suzuki A, Terai H, et al. Time course of physical and mental well-being improvements after cervical surgery. Spine (Phila Pa 1976). 2021;46:E303–9.
Rodrigues-Pinto R, Montenegro TS, Davies BM, et al. Optimizing the application of surgery for degenerative cervical myelopathy [AO Spine RECODE-DCM Research Priority Number 10]. Global Spine J. 2022;12:147S–158S.
Tetreault L, Kalsi-Ryan S, Benjamin D, et al. Degenerative cervical myelopathy: a practical approach to diagnosis. Global Spine J. 2022;12:1881–93.
Nouri AMG, Gondar R, Lavé A, et al. Can screening for degenerative cervical myelopathy (SCREEN-DCM) be effectively undertaken based on signs, symptoms and known risk factors? Rationale and research protocol for a prospective, multicentre, observational study. BMJ Open. 2022;12:e060689.
Terai H, Tamai K, Takahashi S, et al. The health-related quality of life of patients with musculoskeletal disorders after the COVID-19 pandemic. Int Orthop. 2022;46:189–195.
Costanzo G, Misaggi B, Ricciardi L, et al. SPINE20 recommendations 2021: spine care for people’s health and prosperity. Eur Spine J. 2022;31:1333–42.
AlEissa SI, Tamai K, Konbaz F, et al. SPINE20 A global advocacy group promoting evidence-based spine care of value. Eur Spine J. 2021;30:2091–101.
Piche J, Butt BB, Ahmady A, et al. Physical examination of the spine using telemedicine: a systematic review. Global Spine J. 2021;11:1142–7.
Fuson RL, Sherman M, Van Vleet J, et al. The conduct of orthopaedic clinical trials. J Bone Joint Surg Am. 1997;79:1089–98.
Kang Y, Lee JW, Koh YH, et al. New MRI grading system for the cervical canal stenosis. AJR Am J Roentgenol. 2011;197:W134–40.
Tan M, Le QV. EfficientNet: rethinking model scaling for convolutional neural networks. arXiv. 2019;1905:11946.
Kohavi R. A study of cross-validation and bootstrap for accuracy estimation and model selection. IJCAI. 1995;2:1137–43.
Merali Z, Wang JZ, Badhiwala JH, et al. A deep learning model for detection of cervical spinal cord compression in MRI scans. Sci Rep. 2021;11:10473.
Jin R, Luk KD, Cheung JPY, et al. Prognosis of cervical myelopathy based on diffusion tensor imaging with artificial intelligence methods. NMR Biomed. 2019;32:e4114.
Koyama T, Fujita K, Watanabe M, et al. Cervical myelopathy screening with machine learning algorithm focusing on finger motion using noncontact sensor. Spine (Phila Pa 1976). 2022;47:163–71.
Tamai K, Terai H, Hoshino M, et al. A deep learning algorithm to identify cervical ossification of posterior longitudinal ligaments on radiography. Sci Rep. 2022;12:2113.
Gillies RJ, Kinahan PE, Hricak H. Radiomics: images are more than pictures, they are data. Radiology. 2016;278:563–77.
Lee GW, Shin H, Chang MC. Deep learning algorithm to evaluate cervical spondylotic myelopathy using lateral cervical spine radiograph. BMC Neurol. 2022;22:147.
Smith SS, Stewart ME, Davies BM, et al. The prevalence of asymptomatic and symptomatic spinal cord compression on magnetic resonance imaging: a systematic review and meta-analysis. Global Spine J. 2021;11:597–607.
Zhang AS, Myers C, McDonald CL, et al. Cervical myelopathy: diagnosis, contemporary treatment, and outcomes. Am J Med. 2022;135:435–443.
Bajwa NS, Toy JO, Young EY, et al. Establishment of parameters for congenital stenosis of the cervical spine: an anatomic descriptive analysis of 1,066 cadaveric specimens. Eur Spine J. 2012;21:2467–74.
Martin AR, Tetreault L, Nouri A, et al. Imaging and electrophysiology for degenerative cervical myelopathy [AO Spine RECODE-DCM Research Priority Number 9]. Global Spine J. 2022;12:130S–146S.
Minaee S, Boykov YY, Porikli F, et al. Image segmentation using deep learning: a survey. IEEE Trans Pattern Anal Mach Intell. 2020;arXiv:2001.05566v5.
Darwono B, Tamai K, Cote P, et al. SPINE20 recommendations 2022: spine care-working together to recover stronger. Eur Spine J. 2022;31:3262–3273.
Wu E, Wu K, Daneshjou R, et al. How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals. Nat Med. 2021;27:582–4.
Murone I. The importance of the sagittal diameters of the cervical spinal canal in relation to spondylosis and myelopathy. J Bone Joint Surg Br. 1974;56:30–6.
Burrows EH. The sagittal diameter of the spinal canal in cervical spondylosis. Clin Radiol. 1963;14:77–86.