Detection of rheumatoid arthritis from hand radiographs using a convolutional neural network
- 153 Downloads
Plain hand radiographs are the first-line and most commonly used imaging methods for diagnosis or differential diagnosis of rheumatoid arthritis (RA) and for monitoring disease activity. In this study, we used plain hand radiographs and tried to develop an automated diagnostic method using the convolutional neural networks to help physicians while diagnosing rheumatoid arthritis.
A convolutional neural network (CNN) is a deep learning method based on a multilayer neural network structure. The network was trained on a dataset containing 135 radiographs of the right hands, of which 61 were normal and 74 RA, and tested it on 45 radiographs, of which 20 were normal and 25 RA.
The accuracy of the network was 73.33% and the error rate 0.0167. The sensitivity of the network was 0.6818; the specificity was 0.7826 and the precision 0.7500.
Using only pixel information on hand radiographs, a multi-layer CNN architecture with online data augmentation was designed. The performance metrics such as accuracy, error rate, sensitivity, specificity, and precision state shows that the network is promising in diagnosing rheumatoid arthritis.
KeywordsConvolutional neural network Deep learning Plain hand radiographs Rheumatoid arthritis
Compliance with ethical standards
Ethical approval of this study was obtained from the Kırıkkale University Non-Interventional Research Ethics Committee on Dec 2, 2018, no. 2018.09.02.
- 5.Sharp JT, Young DY, Bluhm GB, Brook A, Brower AC, Corbett M, Decker JL, Genant HK, Gofton JP, Goodman N (1985) How many joints in the hands and wrists should be included in a score of radiologic abnormalities used to assess rheumatoid arthritis? Arthritis Rheum 28(12):1326–1335CrossRefPubMedGoogle Scholar
- 6.Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO III, Birnbaum NS, Burmester GR, Bykerk VP, Cohen MD (2010) 2010 rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum 62(9):2569–2581CrossRefGoogle Scholar
- 9.Smolen JS, Landewe R, Breedveld FC, Buch M, Burmester G, Dougados M, Emery P, Gaujoux-Viala C, Gossec L, Nam J, Ramiro S, Winthrop K, de Wit M, Aletaha D, Betteridge N, Bijlsma JW, Boers M, Buttgereit F, Combe B, Cutolo M, Damjanov N, Hazes JM, Kouloumas M, Kvien TK, Mariette X, Pavelka K, van Riel PL, Rubbert-Roth A, Scholte-Voshaar M, Scott DL, Sokka-Isler T, Wong JB, van der Heijde D (2014) EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update. Ann Rheum Dis 73(3):492–509. https://doi.org/10.1136/annrheumdis-2013-204573 CrossRefGoogle Scholar
- 12.Koh JH, Jung SM, Lee JJ, Kang KY, Kwok SK, Park SH, Ju JH (2015) Radiographic structural damage is worse in the dominant than the non-dominant hand in individuals with early rheumatoid arthritis. PLoS One 10(8):e0135409. https://doi.org/10.1371/journal.pone.0135409 CrossRefPubMedPubMedCentralGoogle Scholar
- 15.Banerjee S, Bhunia S, Schaefer G (2011) Osteophyte detection for hand osteoarthritis identification in X-ray images using CNNs. In: Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE. IEEE, pp 6196–6199Google Scholar