AUTHOR=Yeoh Pauline Shan Qing , Lai Khin Wee , Goh Siew Li , Hasikin Khairunnisa , Wu Xiang , Li Pei TITLE=Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative JOURNAL=Frontiers in Bioengineering and Biotechnology VOLUME=Volume 11 - 2023 YEAR=2023 URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2023.1164655 DOI=10.3389/fbioe.2023.1164655 ISSN=2296-4185 ABSTRACT=Knee osteoarthritis is one of the most common musculoskeletal disease and is usually diagnosed with medical imaging techniques. Conventionally, case identification using plain radiography is practiced. However, we acknowledge that knee osteoarthritis is a 3D complexity, hence Magnetic Resonance Imaging will be the ideal modality to reveal the hidden osteoarthritis features from three-dimensional view. In this work, the feasibility of well-known CNN structures (ResNet, DenseNet, VGG, and AlexNet) to distinguish knees with and without OA is investigated. Using 3D convolutional layers, we demonstrated the potential of 3D convolutional neural networks of 13 different architectures in knee osteoarthritis diagnosis. We utilized transfer learning by transforming 2D pretrained weights to 3D as initial weights for the training of the 3D models. The performance of the models was compared and evaluated based on the performance metrics (accuracy, precision, F1 score and area under receiver operating characteristic (AUC) curve). This study suggested that transfer learning indeed enhanced the performance of the models, especially for ResNet and DenseNet models. Transfer learning-based models presented promising results with ResNet34 achieving the best overall accuracy of 0.875 and F1 score of 0.871. The results also showed that shallow networks yielded better performance compared to deeper neural networks, demonstrated by ResNet18, DenseNet121 and VGG11 with AUC of 0.945, 0.914 and 0.928 respectively. This encourages the application of clinical diagnostic aid for knee osteoarthritis using 3DCNN even in limited hardware conditions.