A Lightweight CNN and Joint Shape-Joint Space (JS2) Descriptor for Radiological Osteoarthritis Detection

Knee osteoarthritis (OA) is very common progressive and degenerative musculoskeletal disease worldwide creates a heavy burden on patients with reduced quality of life and also on society due to financial impact. Therefore, any attempt to reduce the burden of the disease could help both patients and society. In this study, we propose a fully automated novel method, based on combination of joint shape and convolutional neural network (CNN) based bone texture features, to distinguish between the knee radiographs with and without radiographic osteoarthritis. Moreover, we report the first attempt at describing the bone texture using CNN. Knee radiographs from Osteoarthritis Initiative (OAI) and Multicenter Osteoarthritis (MOST) studies were used in the experiments. Our models were trained on 8953 knee radiographs from OAI and evaluated on 3445 knee radiographs from MOST. Our results demonstrate that fusing the proposed shape and texture parameters achieves the state-of-the art performance in radiographic OA detection yielding area under the ROC curve (AUC) of 95.21%

[1]  M. Hochberg,et al.  Erosive Osteoarthritis , 2010 .

[2]  Noel E. O'Connor,et al.  Assessing Knee OA Severity with CNN attention-based end-to-end architectures , 2018, MIDL.

[3]  Noel E. O'Connor,et al.  Predicting knee osteoarthritis severity: comparative modeling based on patient’s data and plain X-ray images , 2019, Scientific Reports.

[4]  Timothy F. Cootes,et al.  Automated Shape and Texture Analysis for Detection of Osteoarthritis from Radiographs of the Knee , 2015, MICCAI.

[5]  Thomas M. Link,et al.  Applying Densely Connected Convolutional Neural Networks for Staging Osteoarthritis Severity from Plain Radiographs , 2018, Journal of Digital Imaging.

[6]  Narges Shayesteh Moghaddam,et al.  Achieving superelasticity in additively manufactured NiTi in compression without post-process heat treatment , 2019, Scientific Reports.

[7]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[8]  Pybus Pk,et al.  Osteoarthrosis of the knee. , 1980 .

[9]  Rachid Jennane,et al.  ROI impact on the characterization of knee osteoarthritis using fractal analysis , 2015, 2015 International Conference on Image Processing Theory, Tools and Applications (IPTA).

[10]  D J Hawkes,et al.  A robust and accurate method for calculating the fractal signature of texture in macroradiographs of osteoarthritic knees. , 1991, Medical informatics = Medecine et informatique.

[11]  A. Guermazi,et al.  Plain radiography and magnetic resonance imaging diagnostics in osteoarthritis: validated staging and scoring. , 2009, The Journal of bone and joint surgery. American volume.

[12]  E. Lespessailles,et al.  Subchondral tibial bone texture predicts the incidence of radiographic knee osteoarthritis: data from the Osteoarthritis Initiative. , 2017, Osteoarthritis and cartilage.

[13]  Subhransu Maji,et al.  Improved Bilinear Pooling with CNNs , 2017, BMVC.

[14]  P. J. Murray,et al.  Subchondral bone in osteoarthritis: association between MRI texture analysis and histomorphometry. , 2017, Osteoarthritis and cartilage.

[15]  B. Yılmaz,et al.  Inter- and Intraobserver Reliabilities of Four Different Radiographic Grading Scales of Osteoarthritis of the Knee Joint , 2017, The Journal of Knee Surgery.

[16]  Timothy F. Cootes,et al.  Fully Automatic Segmentation of the Proximal Femur Using Random Forest Regression Voting , 2013, IEEE Transactions on Medical Imaging.

[17]  G. Badger,et al.  Comparison of 2 Radiographic Techniques for Measurement of Tibiofemoral Joint Space Width , 2017, Orthopaedic journal of sports medicine.

[18]  J. Kellgren,et al.  Radiological Assessment of Osteo-Arthrosis , 1957, Annals of the rheumatic diseases.

[19]  Simo Saarakkala,et al.  Automatic Knee Osteoarthritis Diagnosis from Plain Radiographs: A Deep Learning-Based Approach , 2017, Scientific Reports.

[20]  R. Altman.  Early management of osteoarthritis. , 2010, The American journal of managed care.

[21]  Simo Saarakkala,et al.  Adaptive Segmentation of Knee Radiographs for Selecting the Optimal ROI in Texture Analysis , 2019, Osteoarthritis and cartilage.

[22]  U. Wyss,et al.  Trabecular microstructure in the medial condyle of the proximal tibia of patients with knee osteoarthritis. , 1995, Bone.

[23]  C. Buckland-Wright.  Subchondral bone changes in hand and knee osteoarthritis detected by radiography. , 2004, Osteoarthritis and cartilage.

[24]  M. Nevitt,et al.  Predictive Validity of Radiographic Trabecular Bone Texture in Knee Osteoarthritis , 2017, Arthritis & rheumatology.

[25]  F. Berenbaum,et al.  Osteoarthritis: an update with relevance for clinical practice , 2011, The Lancet.

[26]  H. Genant,et al.  New radiographic-based surrogate outcome measures for osteoarthritis of the knee. , 2003, Osteoarthritis and cartilage.

[27]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[28]  Joseph Antony,et al.  Quantifying radiographic knee osteoarthritis severity using deep convolutional neural networks , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).

[29]  D. Hawkes,et al.  Analysis of texture in macroradiographs of osteoarthritic knees using the fractal signature. , 1991, Physics in medicine and biology.

[30]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[31]  P. Cao,et al.  Discovering knee osteoarthritis bone shape features using deep learning , 2019, Osteoarthritis and Cartilage.

[32]  R. Aspden,et al.  Osteoarthritis as an organ disease: from the cradle to the grave. , 2019, European cells & materials.

[33]  M. Nevitt,et al.  Location specific radiographic joint space width for osteoarthritis progression. , 2009, Osteoarthritis and cartilage.

[34]  P. K. Pybus,et al.  Osteoarthrosis of the knee. , 1980, The Practitioner.

[35]  S. Bierma-Zeinstra,et al.  Variation in joint shape of osteoarthritic knees. , 2011, Arthritis and rheumatism.

[36]  Timothy F. Cootes,et al.  Fully automated shape analysis for detection of Osteoarthritis from lateral knee radiographs , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).

[37]  Matthias Bethge,et al.  ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness , 2018, ICLR.

[38]  B. Heidari.  Knee osteoarthritis prevalence, risk factors, pathogenesis and features: Part I. , 2011, Caspian journal of internal medicine.

[39]  Matti Pietikäinen,et al.  Gray Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2000, ECCV.

[40]  Richard Zhang,et al.  Making Convolutional Networks Shift-Invariant Again , 2019, ICML.

[41]  J. Arokoski,et al.  Quantification of differences in bone texture from plain radiographs in knees with and without osteoarthritis. , 2014, Osteoarthritis and cartilage.

[42]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[43]  S. Goldring.  Role of bone in osteoarthritis pathogenesis. , 2009, The Medical clinics of North America.

[44]  H. Kokkonen,et al.  Correlation of Subchondral Bone Density and Structure from Plain Radiographs with Micro Computed Tomography Ex Vivo , 2015, Annals of Biomedical Engineering.