Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Sep 29;12(10):2362.
doi: 10.3390/diagnostics12102362.

Detection and Classification of Knee Osteoarthritis

Joseph Humberto Cueva  1 , Darwin Castillo  1   2   3 , Héctor Espinós-Morató  4 , David Durán  5 , Patricia Díaz  6 , Vasudevan Lakshminarayanan  3   7
Affiliations

Detection and Classification of Knee Osteoarthritis

Joseph Humberto Cueva et al. Diagnostics (Basel). .

Abstract

Osteoarthritis (OA) affects nearly 240 million people worldwide. Knee OA is the most common type of arthritis, especially in older adults. Physicians measure the severity of knee OA according to the Kellgren and Lawrence (KL) scale through visual inspection of X-ray or MR images. We propose a semi-automatic CADx model based on Deep Siamese convolutional neural networks and a fine-tuned ResNet-34 to simultaneously detect OA lesions in the two knees according to the KL scale. The training was done using a public dataset, whereas the validations were performed with a private dataset. Some problems of the imbalanced dataset were solved using transfer learning. The model results average of the multi-class accuracy is 61%, presenting better performance results for classifying classes KL-0, KL-3, and KL-4 than KL-1 and KL-2. The classification results were compared and validated using the classification of experienced radiologists.

Keywords: CAD; CNN; KL grades; OA knee; X-ray images; deep learning; osteoarthritis.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Representation of the ResNet-34 architecture modified with two fully connected (FC) layers added. The figure was adapted from [28].
Figure 2
Figure 2
Representation of the Siamese network architecture. The lateral side of the knee X-ray image feeds one branch of the network, and the medial side of the knee X-ray image provides the second branch of the network. The figure was adapted from [28].
Figure 3
Figure 3
Software pipeline used in this project. The figure was adapted from [29].
Figure 4
Figure 4
Confusion matrix for KL grading according to the proposed model.
Figure 5
Figure 5
Evolution of the accuracy for the training set and the validation set.
Figure 6
Figure 6
Comparison with the experts. The yellow bar represents the classification results of the proposed model. There is a low classification accuracy for KL-1 and KL-2 grades.
Figure 7
Figure 7
(a) Navigation panel, that allows selection of the image to process (*.dcm files). (b) The X-ray image panel where the region of interest is selected.
Figure 8
Figure 8
Result panel where the probabilities of the classification according to KL grades in each knee of the patient are shown.
Figure 9
Figure 9
The graphic interface where the classification of severity of the knee osteoarthritis (KOA) is shown according to the KL grades from the radiographic image.
See this image and copyright information in PMC

References

    1. Nelson A.E. Osteoarthritis Year in Review 2017: Clinical. Osteoarthr. Cartil. 2018;26:319–325. doi: 10.1016/j.joca.2017.11.014. - DOI - PMC - PubMed
    1. Tiulpin A., Thevenot J., Rahtu E., Lehenkari P., Saarakkala S. Automatic Knee Osteoarthritis Diagnosis from Plain Radiographs: A Deep Learning-Based Approach. Sci. Rep. 2018;8:1727. doi: 10.1038/s41598-018-20132-7. - DOI - PMC - PubMed
    1. Chen P., Gao L., Shi X., Allen K., Yang L. Fully Automatic Knee Osteoarthritis Severity Grading Using Deep Neural Networks with a Novel Ordinal Loss. Comput. Med. Imaging Graph. 2019;75:84–92. doi: 10.1016/j.compmedimag.2019.06.002. - DOI - PMC - PubMed
    1. Abedin J., Antony J., McGuinness K., Moran K., O’Connor N.E., Rebholz-Schuhmann D., Newell J. Predicting Knee Osteoarthritis Severity: Comparative Modeling Based on Patient’s Data and Plain X-Ray Images. Sci. Rep. 2019;9:5761. doi: 10.1038/s41598-019-42215-9. - DOI - PMC - PubMed
    1. Kalo K., Niederer D., Schmitt M., Vogt L. Acute effects of a single bout of exercise therapy on knee acoustic emissions in patients with osteoarthritis: A double-blinded, randomized controlled crossover trial. BMC Musculoskelet. Disord. 2022;23:657. doi: 10.1186/s12891-022-05616-y. - DOI - PMC - PubMed

LinkOut - more resources