Abstract
Purpose
The aim of this work is to propose a classification algorithm to automatically detect treatment for scoliosis (brace, implant or no treatment) in postero-anterior radiographs. Such automatic labelling of radiographs could represent a step towards global automatic radiological analysis.
Methods
Seven hundred and ninety-six frontal radiographies of adolescents were collected (84 patients wearing a brace, 325 with a spinal implant and 387 reference images with no treatment). The dataset was augmented to a total of 2096 images. A classification model was built, composed by a forward convolutional neural network (CNN) followed by a discriminant analysis; the output was a probability for a given image to contain a brace, a spinal implant or none. The model was validated with a stratified tenfold cross-validation procedure. Performance was estimated by calculating the average accuracy.
Results
98.3% of the radiographs were correctly classified as either reference, brace or implant, excluding 2.0% unclassified images. 99.7% of brace radiographs were correctly detected, while most of the errors occurred in the reference group (i.e. 2.1% of reference images were wrongly classified).
Conclusion
The proposed classification model, the originality of which is the coupling of a CNN with discriminant analysis, can be used to automatically label radiographs for the presence of scoliosis treatment. This information is usually missing from DICOM metadata, so such method could facilitate the use of large databases. Furthermore, the same model architecture could potentially be applied for other radiograph classifications, such as sex and presence of scoliotic deformity.
References
Cheng JC, Castelein RM, Chu WC, Danielsson AJ, Dobbs MB, Grivas TB, Gurnett CA, Luk KD, Moreau A, Newton PO, Stokes IA, Weinstein SL, Burwell RG (2015) Adolescent idiopathic scoliosis. Nat Rev Dis Primers 1:15030. https://doi.org/10.1038/nrdp.2015.30
Negrini S, Donzelli S, Aulisa AG, Czaprowski D, Schreiber S, de Mauroy JC, Diers H, Grivas TB, Knott P, Kotwicki T, Lebel A, Marti C, Maruyama T, O’Brien J, Price N, Parent E, Rigo M, Romano M, Stikeleather L, Wynne J, Zaina F (2018) 2016 SOSORT guidelines: orthopaedic and rehabilitation treatment of idiopathic scoliosis during growth. Scoliosis Spinal Disord 13:3. https://doi.org/10.1186/s13013-017-0145-8
Weinstein SL, Dolan LA, Wright JG, Dobbs MB (2013) Effects of bracing in adolescents with idiopathic scoliosis. N Engl J Med 369:1512–1521. https://doi.org/10.1056/NEJMoa1307337
Bunnell WP (1986) The natural history of idiopathic scoliosis before skeletal maturity. Spine (Phila Pa 1976) 11:773–776
Galbusera F, Casaroli G, Bassani T (2019) Artificial intelligence and machine learning in spine research. Jor Spine 2:e1044. https://doi.org/10.1002/jsp2.1044
Bakhous C, Aubert B, Vazquez C, Cresson T, Parent S, Guise J De (2018) Automatic pedicles detection using convolutional neural network in a 3D spine reconstruction from biplanar radiographs. In: Proc SPIE
Ebrahimi S, Gajny L, Skalli W, Angelini E (2019) Vertebral corners detection on sagittal X-rays based on shape modelling, random forest classifiers and dedicated visual features. Comput Methods Biomech Biomed Eng Imaging Vis 7:132–144. https://doi.org/10.1080/21681163.2018.1463174
Aubert B, Vazquez C, Cresson T, Parent S, de Guise JA (2019) Toward automated 3D spine reconstruction from biplanar radiographs using CNN for statistical spine model fitting. IEEE Trans Med Imaging 38:2796–2806. https://doi.org/10.1109/TMI.2019.2914400
Yang Z, Skalli W, Vergari C, Angelini ED, Gajny L (2019) Automated spinal midline delineation on biplanar x-rays using mask R-CNN - VipIMAGE. In: Natal Jorge RM (ed) Tavares JMRS. Springer International Publishing, Cham, pp 307–316
Zhang J, Li H, Lv L, Zhang Y (2017) Computer-aided Cobb measurement based on automatic detection of vertebral slopes using deep neural network. Int J Biomed Imaging 2017:9083916. https://doi.org/10.1155/2017/9083916
Cho BH, Kaji D, Cheung ZB, Ye IB, Tang R, Ahn A, Carrillo O, Schwartz JT, Valliani AA, Oermann EK, Arvind V, Ranti D, Sun L, Kim JS, Cho SK (2019) Automated measurement of lumbar lordosis on radiographs using machine learning and computer vision. Glob Spine J. https://doi.org/10.1177/2192568219868190
Weng C-H, Wang C-L, Huang Y-J, Yeh Y-C, Fu C-J, Yeh C-Y, Tsai T-T (2019) Artificial intelligence for automatic measurement of sagittal vertical axis using ResUNet framework. J Clin Med 8(11):1826
Klinder T, Ostermann J, Ehm M, Franz A, Kneser R, Lorenz C (2009) Automated model-based vertebra detection, identification, and segmentation in CT images. Med Image Anal 13:471–482. https://doi.org/10.1016/j.media.2009.02.004
Zheng G, Chu C, Belavý DL, Ibragimov B, Korez R, Vrtovec T, Hutt H, Everson R, Meakin J, Andrade IL, Glocker B, Chen H, Dou Q, Heng P-A, Wang C, Forsberg D, Neubert A, Fripp J, Urschler M, Stern D, Wimmer M, Novikov A, Cheng H, Armbrecht G, Felsenberg D, Li S (2017) Evaluation and comparison of 3D intervertebral disc localization and segmentation methods for 3D T2 MR data: a grand challenge. Med Image Anal 35:327–344. https://doi.org/10.1016/j.media.2016.08.005
Jamaludin A, Lootus M, Kadir T, Zisserman A, Urban J, Battié MC, Fairbank J, McCall I (2017) ISSLS PRIZE IN BIOENGINEERING SCIENCE 2017: automation of reading of radiological features from magnetic resonance images (MRIs) of the lumbar spine without human intervention is comparable with an expert radiologist. Eur Spine J 26:1374–1383. https://doi.org/10.1007/s00586-017-4956-3
Faro FD, Marks MC, Pawelek J, Newton PO (2004) Evaluation of a functional position for lateral radiograph acquisition in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 29:2284–2289
Lecun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86:2278–2324. https://doi.org/10.1109/5.726791
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15:1929–1958
Vedaldi A, Lenc K (2015) MatConvNet—convolutional neural networks for MATLAB. In: Proceeding of the ACM Int Conf on Multimedia
Pedersen PH, Vergari C, Alzakri A, Vialle R, Skalli W (2019) A reduced micro-dose protocol for 3D reconstruction of the spine in children with scoliosis: results of a phantom-based and clinically validated study using stereo-radiography. Eur Radiol 29:1874–1881. https://doi.org/10.1007/s00330-018-5749-8
Dubousset J, Ilharreborde B, Le Huec J-C (2014) Use of EOS imaging for the assessment of scoliosis deformities: application to postoperative 3D quantitative analysis of the trunk. Eur Spine J 23:397–405. https://doi.org/10.1007/s00586-014-3334-7
Vergari C, Aubert B, Lallemant-Dudek P, Haen TX, Skalli W (2019) A novel method of anatomical landmark selection for rib cage 3D reconstruction from biplanar radiography. Comput Methods Biomech Biomed Eng Imaging Vis. https://doi.org/10.1080/21681163.2018.1537860
Hirsch C, Ilharreborde B, Mazda K (2016) Flexibility analysis in adolescent idiopathic scoliosis on side-bending images using the EOS imaging system. Orthop Traumatol Surg Res 102:495–500. https://doi.org/10.1016/j.otsr.2016.01.021
Acknowledgements
The authors are grateful to the ParisTech BiomecAM chair program on subject-specific musculoskeletal modelling (with the support of ParisTech and Yves Cotrel Foundations, Société Générale, Proteor and Covea).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Wafa Skalli holds patents related to the EOS system and associated 3D reconstruction methods, with no personal financial benefit (royalties rewarded for research and education). The other authors declare that they have no conflict of interest.
Ethical standards
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Informed consent
For this type of study, formal consent is not required.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Vergari, C., Skalli, W. & Gajny, L. A convolutional neural network to detect scoliosis treatment in radiographs. Int J CARS 15, 1069–1074 (2020). https://doi.org/10.1007/s11548-020-02173-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11548-020-02173-4