Automatic Grading of Individual Knee Osteoarthritis Features in Plain Radiographs Using Deep Convolutional Neural Networks

Aleksei Tiulpin; Simo Saarakkala

doi:10.3390/diagnostics10110932

journal article Open Access Nov 10, 2020

Automatic Grading of Individual Knee Osteoarthritis Features in Plain Radiographs Using Deep Convolutional Neural Networks

Aleksei Tiulpin

Simo Saarakkala

Diagnostics Vol. 10 No. 11 pp. 932 · MDPI AG

View at Publisher Save 10.3390/diagnostics10110932

Abstract

Knee osteoarthritis (OA) is the most common musculoskeletal disease in the world. In primary healthcare, knee OA is diagnosed using clinical examination and radiographic assessment. Osteoarthritis Research Society International (OARSI) atlas of OA radiographic features allows performing independent assessment of knee osteophytes, joint space narrowing and other knee features. This provides a fine-grained OA severity assessment of the knee, compared to the gold standard and most commonly used Kellgren–Lawrence (KL) composite score. In this study, we developed an automatic method to predict KL and OARSI grades from knee radiographs. Our method is based on Deep Learning and leverages an ensemble of residual networks with 50 layers. We used transfer learning from ImageNet with a fine-tuning on the Osteoarthritis Initiative (OAI) dataset. An independent testing of our model was performed on the Multicenter Osteoarthritis Study (MOST) dataset. Our method yielded Cohen’s kappa coefficients of 0.82 for KL-grade and 0.79, 0.84, 0.94, 0.83, 0.84 and 0.90 for femoral osteophytes, tibial osteophytes and joint space narrowing for lateral and medial compartments, respectively. Furthermore, our method yielded area under the ROC curve of 0.98 and average precision of 0.98 for detecting the presence of radiographic OA, which is better than the current state-of-the-art.

Topics

No keywords indexed for this article. Browse by subject →

References

33

[1]

Arden "Osteoarthritis: Epidemiology" Best Pract. Res. Clin. Rheumatol. (2006) 10.1016/j.berh.2005.09.007

[2]

Cross "The global burden of hip and knee osteoarthritis: Estimates from the global burden of disease 2010 study" Ann. Rheum. Dis. (2014) 10.1136/annrheumdis-2013-204763

[3]

Wluka "Tackling obesity in knee osteoarthritis" Nat. Rev. Rheumatol. (2013) 10.1038/nrrheum.2012.224

[4]

Tiulpin "Automatic knee osteoarthritis diagnosis from plain radiographs: A deep learning-based approach" Sci. Rep. (2018) 10.1038/s41598-018-20132-7

[5]

Radiological Assessment of Osteo-Arthrosis

J.H. Kellgren, J.S. Lawrence

Annals of the Rheumatic Diseases 1957 10.1136/ard.16.4.494

[6]

Atlas of individual radiographic features in osteoarthritis, revised

R.D. Altman, G.E. Gold

Osteoarthritis and Cartilage 2007 10.1016/j.joca.2006.11.009

[7]

Esteva "A guide to deep learning in healthcare" Nat. Med. (2019) 10.1038/s41591-018-0316-z

[8]

Pedoia "3D convolutional neural networks for detection and severity staging of meniscus and PFJ cartilage morphological degenerative changes in osteoarthritis and anterior cruciate ligament subjects" J. Magn. Reson. Imaging (2019) 10.1002/jmri.26246

[9]

Norman "Use of 2D U-Net convolutional neural networks for automated cartilage and meniscus segmentation of knee MR imaging data to determine relaxometry and morphometry" Radiology (2018) 10.1148/radiol.2018172322

[10]

Tiulpin, A., Finnilä, M., Lehenkari, P., Nieminen, H.J., and Saarakkala, S. (2019). Deep-Learning for Tidemark Segmentation in Human Osteochondral Tissues Imaged with Micro-computed Tomography. arXiv. 10.1007/978-3-030-40605-9_12

[11]

Tiulpin, A., Klein, S., Bierma-Zeinstra, S., Thevenot, J., Rahtu, E., Van Meurs, J., Oei, E.H., and Saarakkala, S. (2019). Multimodal Machine Learning-based Knee Osteoarthritis Progression Prediction from Plain Radiographs and Clinical Data. arXiv. 10.1038/s41598-019-56527-3

[12]

Antony, J., McGuinness, K., Moran, K., and O’Connor, N.E. (2017). Automatic detection of knee joints and quantification of knee osteoarthritis severity using convolutional neural networks. Proceedings of the International Conference on Machine Learning and Data Mining in Pattern Recognition, Leipzig, Germany, 18–20 July 2007, Springer. 10.1007/978-3-319-62416-7_27

[13]

Norman "Applying Densely Connected Convolutional Neural Networks for Staging Osteoarthritis Severity from Plain Radiographs" J. Digit. Imaging (2018) 10.1007/s10278-018-0098-3

[14]

Xue, Y., Zhang, R., Deng, Y., Chen, K., and Jiang, T. (2017). A preliminary examination of the diagnostic value of deep learning in hip osteoarthritis. PLoS ONE, 12. 10.1371/journal.pone.0178992

[15]

Antony, A.J. (2018). Automatic Quantification of Radiographic Knee Osteoarthritis Severity and Associated Diagnostic Features Using Deep Convolutional Neural Networks. [Ph.D. Thesis, Dublin City University].

[16]

Squeeze-and-Excitation Networks

Jie Hu, Li Shen, Gang Sun

2018 IEEE/CVF Conference on Computer Vision and Pa... 10.1109/cvpr.2018.00745

[17]

Aggregated Residual Transformations for Deep Neural Networks

Saining Xie, Ross Girshick, Piotr Dollar et al.

2017 IEEE Conference on Computer Vision and Patter... 10.1109/cvpr.2017.634

[18]

Lindner "Fully automatic segmentation of the proximal femur using random forest regression voting" IEEE Trans. Med. Imaging (2013) 10.1109/tmi.2013.2258030

[19]

Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning

Hoo-Chang Shin, Holger R. Roth, Mingchen Gao et al.

IEEE Transactions on Medical Imaging 2016 10.1109/tmi.2016.2528162

[20]

ImageNet: A large-scale hierarchical image database

Jia Deng, Wei Dong, Richard Socher et al.

2009 IEEE Conference on Computer Vision and Patter... 10.1109/cvpr.2009.5206848

[21]

Kothari "Fixed-flexion radiography of the knee provides reproducible joint space width measurements in osteoarthritis" Eur. Radiol. (2004) 10.1007/s00330-004-2312-6

[22]

Tiulpin, A., Thevenot, J., Rahtu, E., and Saarakkala, S. (2017). A Novel Method for Automatic Localization of Joint Area on Knee Plain Radiographs. Proceedings of the Scandinavian Conference on Image Analysis, Tromsø, Norway, 12–14 June 2017, Springer. 10.1007/978-3-319-59129-2_25

[23]

Deep Residual Learning for Image Recognition

Kaiming He, Xiangyu Zhang, Shaoqing Ren et al.

2016 IEEE Conference on Computer Vision and Patter... 10.1109/cvpr.2016.90

[24]

Qiu, S. (2018). Global Weighted Average Pooling Bridges Pixel-level Localization and Image-level Classification. arXiv.

[25]

Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv.

[26]

Tiulpin, A. (2020, November 10). SOLT: Streaming over Lightweight Transformations. Available online: https://github.com/MIPT-Oulu/solt.

[27]

Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., and Lerer, A. (2020, November 10). Automatic Differentiation in PyTorch. NIPS-W. Available online: https://openreview.net/forum?id=BJJsrmfCZ.

[28]

Riddle "Validity and reliability of radiographic knee osteoarthritis measures by arthroplasty surgeons" Orthopedics (2013) 10.3928/01477447-20121217-14

[29]

Oka "Normal and threshold values of radiographic parameters for knee osteoarthritis using a computer-assisted measuring system (KOACAD): The ROAD study" J. Orthop. Sci. (2010) 10.1007/s00776-010-1545-2

[30]

Thomson, J., O’Neill, T., Felson, D., and Cootes, T. (2016). Detecting Osteophytes in Radiographs of the Knee to Diagnose Osteoarthritis. Proceedings of the International Workshop on Machine Learning in Medical Imaging, Athens, Greece, 17 October 2016, Springer. 10.1007/978-3-319-47157-0_6

[31]

Quantifying radiographic knee osteoarthritis severity using deep convolutional neural networks

Joseph Antony, Kevin McGuinness, Noel E O'Connor et al.

2016 23rd International Conference on Pattern Reco... 10.1109/icpr.2016.7899799

[32]

Opportunities and obstacles for deep learning in biology and medicine

Travers Ching, Daniel S. Himmelstein, Brett K. Beaulieu-Jones et al.

Journal of The Royal Society Interface 2018 10.1098/rsif.2017.0387

[33]

Hinton, G., Vinyals, O., and Dean, J. (2015). Distilling the knowledge in a neural network. arXiv.

Cited By

100

From variability to consistency: building a Kellgren-Lawrence gonarthrosis dataset

Salih Beyaz, Sahika Betul Yayli · 2025

Journal of Orthopaedic Surgery and...

Bone Segmentation in Low-Field Knee MRI Using a Three-Dimensional Convolutional Neural Network

Ciro Listone, Diego Romano · 2025

Big Data and Cognitive Computing

A Comparative Study of Machine Learning Classifiers for Enhancing Knee Osteoarthritis Diagnosis

Aquib Raza, Thien-Luan Phan · 2024

Information

Ensemble deep-learning networks for automated osteoarthritis grading in knee X-ray images

Sun-Woo Pi, Byoung-Dai Lee · 2023

Scientific Reports

Automating classification of osteoarthritis according to Kellgren-Lawrence in the knee using deep learning in an unfiltered adult population

Simon Olsson, Ehsan Akbarian · 2021

BMC Musculoskeletal Disorders

Metrics

100

Citations

33

References

Details

Published: Nov 10, 2020
Vol/Issue: 10(11)
Pages: 932
License: View

Authors

A

Aleksei Tiulpin

Research Unit of Medical Imaging, Physics and Technology, University of Oulu, 90220 Oulu, Finland; Department of Diagnostic Radiology, Oulu University Hospital, 90220 Oulu, Finland; Ailean Technologies Oy, 90230 Oulu, Finland

S

Simo Saarakkala

Research Unit of Medical Imaging, Physics and Technology, University of Oulu, 90220 Oulu, Finland; Department of Diagnostic Radiology, Oulu University Hospital, 90220 Oulu, Finland

Funding

Kaute-Säätiö Award: Doctoral research grant 05/2017

Cite This Article

Aleksei Tiulpin, Simo Saarakkala (2020). Automatic Grading of Individual Knee Osteoarthritis Features in Plain Radiographs Using Deep Convolutional Neural Networks. Diagnostics, 10(11), 932. https://doi.org/10.3390/diagnostics10110932

Automatic Grading of Individual Knee Osteoarthritis Features in Plain Radiographs Using Deep Convolutional Neural Networks

You May Also Like