Üreten Kemal, Erbay Hasan, Maraş Hadi Hakan
Department of Rheumatology, Faculty of Medicine, Kırıkkale University, 71450, Kırıkkale, Turkey.
Department of Computer Engineering (MSc), Çankaya University, Ankara, Turkey.
Clin Rheumatol. 2020 Apr;39(4):969-974. doi: 10.1007/s10067-019-04487-4. Epub 2019 Mar 8.
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.
手部X线平片是类风湿关节炎（RA）诊断或鉴别诊断以及监测疾病活动的一线且最常用的影像学方法。在本研究中，我们使用手部X线平片，并尝试开发一种利用卷积神经网络的自动诊断方法，以在诊断类风湿关节炎时帮助医生。
卷积神经网络（CNN）是一种基于多层神经网络结构的深度学习方法。该网络在包含135张右手X线片的数据集上进行训练，其中61张为正常，74张为类风湿关节炎，并用45张X线片进行测试，其中20张为正常，25张为类风湿关节炎。
该网络的准确率为73.33%，错误率为0.0167。该网络的灵敏度为0.6818；特异性为0.7826，精确率为0.7500。
仅利用手部X线片上的像素信息，设计了一种具有在线数据增强功能的多层CNN架构。诸如准确率、错误率、灵敏度、特异性和精确率等性能指标表明，该网络在类风湿关节炎诊断方面具有前景。
