2020/05/24 by Neslihan Bayramoğlu, Bayramoglu, Neslihan, Miika T. Nieminen +3
Medicine · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #Osteoarthritis Treatment and Mechanisms #Rheumatoid Arthritis Research and Therapies #Traditional Chinese Medicine Studies #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2005.11715
openalex publication_date 2020/05/24 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Knee osteoarthritis (OA) is very common progressive and degenerative\nmusculoskeletal disease worldwide creates a heavy burden on patients with\nreduced quality of life and also on society due to financial impact. Therefore,\nany attempt to reduce the burden of the disease could help both patients and\nsociety. In this study, we propose a fully automated novel method, based on\ncombination of joint shape and convolutional neural network (CNN) based bone\ntexture features, to distinguish between the knee radiographs with and without\nradiographic osteoarthritis. Moreover, we report the first attempt at\ndescribing the bone texture using CNN. Knee radiographs from Osteoarthritis\nInitiative (OAI) and Multicenter Osteoarthritis (MOST) studies were used in the\nexperiments. Our models were trained on 8953 knee radiographs from OAI and\nevaluated on 3445 knee radiographs from MOST. Our results demonstrate that\nfusing the proposed shape and texture parameters achieves the state-of-the art\nperformance in radiographic OA detection yielding area under the ROC curve\n(AUC) of 95.21%\n