Cargando…
Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative
Knee osteoarthritis is one of the most common musculoskeletal diseases 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 im...
Autores principales: | , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10136763/ https://www.ncbi.nlm.nih.gov/pubmed/37122858 http://dx.doi.org/10.3389/fbioe.2023.1164655 |
_version_ | 1785032297556213760 |
---|---|
author | Yeoh, Pauline Shan Qing Lai, Khin Wee Goh, Siew Li Hasikin, Khairunnisa Wu, Xiang Li, Pei |
author_facet | Yeoh, Pauline Shan Qing Lai, Khin Wee Goh, Siew Li Hasikin, Khairunnisa Wu, Xiang Li, Pei |
author_sort | Yeoh, Pauline Shan Qing |
collection | PubMed |
description | Knee osteoarthritis is one of the most common musculoskeletal diseases 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 a three-dimensional view. In this work, the feasibility of well-known convolutional neural network (CNN) structures (ResNet, DenseNet, VGG, and AlexNet) to distinguish knees with and without osteoarthritis (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 used transfer learning by transforming 2D pre-trained weights into 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 [balanced 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 an F1 score of 0.871. The results also showed that shallow networks yielded better performance than deeper neural networks, demonstrated by ResNet18, DenseNet121, and VGG11 with AUC values 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. |
format | Online Article Text |
id | pubmed-10136763 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-101367632023-04-28 Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative Yeoh, Pauline Shan Qing Lai, Khin Wee Goh, Siew Li Hasikin, Khairunnisa Wu, Xiang Li, Pei Front Bioeng Biotechnol Bioengineering and Biotechnology Knee osteoarthritis is one of the most common musculoskeletal diseases 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 a three-dimensional view. In this work, the feasibility of well-known convolutional neural network (CNN) structures (ResNet, DenseNet, VGG, and AlexNet) to distinguish knees with and without osteoarthritis (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 used transfer learning by transforming 2D pre-trained weights into 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 [balanced 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 an F1 score of 0.871. The results also showed that shallow networks yielded better performance than deeper neural networks, demonstrated by ResNet18, DenseNet121, and VGG11 with AUC values 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. Frontiers Media S.A. 2023-04-13 /pmc/articles/PMC10136763/ /pubmed/37122858 http://dx.doi.org/10.3389/fbioe.2023.1164655 Text en Copyright © 2023 Yeoh, Lai, Goh, Hasikin, Wu and Li. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
spellingShingle | Bioengineering and Biotechnology Yeoh, Pauline Shan Qing Lai, Khin Wee Goh, Siew Li Hasikin, Khairunnisa Wu, Xiang Li, Pei Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative |
title | Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative |
title_full | Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative |
title_fullStr | Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative |
title_full_unstemmed | Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative |
title_short | Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative |
title_sort | transfer learning-assisted 3d deep learning models for knee osteoarthritis detection: data from the osteoarthritis initiative |
topic | Bioengineering and Biotechnology |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10136763/ https://www.ncbi.nlm.nih.gov/pubmed/37122858 http://dx.doi.org/10.3389/fbioe.2023.1164655 |
work_keys_str_mv | AT yeohpaulineshanqing transferlearningassisted3ddeeplearningmodelsforkneeosteoarthritisdetectiondatafromtheosteoarthritisinitiative AT laikhinwee transferlearningassisted3ddeeplearningmodelsforkneeosteoarthritisdetectiondatafromtheosteoarthritisinitiative AT gohsiewli transferlearningassisted3ddeeplearningmodelsforkneeosteoarthritisdetectiondatafromtheosteoarthritisinitiative AT hasikinkhairunnisa transferlearningassisted3ddeeplearningmodelsforkneeosteoarthritisdetectiondatafromtheosteoarthritisinitiative AT wuxiang transferlearningassisted3ddeeplearningmodelsforkneeosteoarthritisdetectiondatafromtheosteoarthritisinitiative AT lipei transferlearningassisted3ddeeplearningmodelsforkneeosteoarthritisdetectiondatafromtheosteoarthritisinitiative |