Knee osteoarthritis (OA) is a degenerative joint disease that causes physical disability worldwide and has a significant impact on public health. The diagnosis of OA is often made from X-ray images, however, this diagnosis suffers from subjectivity as it is achieved visually by evaluating symptoms according to the radiologist experience/expertise. In this article, we introduce a new deep convolutional neural network based on the standard DenseNet model to automatically score early knee OA from X-ray images. Our method consists of two main ideas: improving network texture analysis to better identify early signs of OA, and combining prediction loss with a novel discriminative loss to address the problem of the high similarity shown between knee joint radiographs of OA and non-OA subjects. Comprehensive experimental results over two large public databases demonstrate the potential of the proposed network.
CITATION STYLE
Nasser, Y., Hassouni, M. E., & Jennane, R. (2022). Discriminative Deep Neural Network for Predicting Knee OsteoArthritis in Early Stage. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 13564 LNCS, pp. 126–136). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-16919-9_12
Mendeley helps you to discover research relevant for your work.