Prediction of Knee Prosthesis Using Patient Gender and BMI With Non-marked X-Ray by Deep Learning.

deep learning error correct output coding prosthesis prediction total knee arthroplasty transfer learning

Journal

Frontiers in surgery
ISSN: 2296-875X
Titre abrégé: Front Surg
Pays: Switzerland
ID NLM: 101645127

Informations de publication

Date de publication:
2022
Historique:
received: 20 10 2021
accepted: 11 02 2022
entrez: 1 4 2022
pubmed: 2 4 2022
medline: 2 4 2022
Statut: epublish

Résumé

Total knee arthroplasty (TKA) is effective for severe osteoarthritis and other related diseases. Accurate prosthesis prediction is a crucial factor for improving clinical outcomes and patient satisfaction after TKA. Current studies mainly focus on conventional manual template measurements, which are inconvenient and inefficient. In this article, we utilize convolutional neural networks to analyze a multimodal patient data and design a system that helps doctors choose prostheses for TKA. To alleviate the problems of insufficient data and uneven distribution of labels, research on model structure, loss function and transfer learning is carried out. Algorithm optimization based on error correct output coding (ECOC) is implemented to further boost the performance. The experimental results show the ECOC-based model reaches prediction accuracies of 88.23% and 86.27% for femoral components and tibial components, respectively. The results verify that the ECOC-based model for prosthesis prediction in TKA is feasible and outperforms existing methods, which is of great significance for templating.

Sections du résumé

Background UNASSIGNED
Total knee arthroplasty (TKA) is effective for severe osteoarthritis and other related diseases. Accurate prosthesis prediction is a crucial factor for improving clinical outcomes and patient satisfaction after TKA. Current studies mainly focus on conventional manual template measurements, which are inconvenient and inefficient.
Methods UNASSIGNED
In this article, we utilize convolutional neural networks to analyze a multimodal patient data and design a system that helps doctors choose prostheses for TKA. To alleviate the problems of insufficient data and uneven distribution of labels, research on model structure, loss function and transfer learning is carried out. Algorithm optimization based on error correct output coding (ECOC) is implemented to further boost the performance.
Results UNASSIGNED
The experimental results show the ECOC-based model reaches prediction accuracies of 88.23% and 86.27% for femoral components and tibial components, respectively.
Conclusions UNASSIGNED
The results verify that the ECOC-based model for prosthesis prediction in TKA is feasible and outperforms existing methods, which is of great significance for templating.

Identifiants

pubmed: 35360429
doi: 10.3389/fsurg.2022.798761
pmc: PMC8963922
doi:

Types de publication

Journal Article

Langues

eng

Pagination

798761

Informations de copyright

Copyright © 2022 Yue, Gao, Zhao, Li and Tian.

Déclaration de conflit d'intérêts

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Références

Annu Int Conf IEEE Eng Med Biol Soc. 2017 Jul;2017:4207-4210
pubmed: 29060825
IEEE Trans Neural Netw Learn Syst. 2014 Feb;25(2):289-302
pubmed: 24807029
Int Orthop. 2014 Apr;38(4):733-9
pubmed: 24162155
Arch Orthop Trauma Surg. 2018 Mar;138(3):409-418
pubmed: 29177951
Orthop Clin North Am. 2017 Oct;48(4):421-431
pubmed: 28870303
J Arthroplasty. 2015 Oct;30(10):1699-704
pubmed: 26021908
J Arthroplasty. 2017 Dec;32(12):3854-3860
pubmed: 28844632
Sci Rep. 2018 Jan 29;8(1):1727
pubmed: 29379060
Knee Surg Sports Traumatol Arthrosc. 2013 Oct;21(10):2314-24
pubmed: 23404515
Nat Med. 2019 Jan;25(1):44-56
pubmed: 30617339
IEEE Trans Nanobioscience. 2018 Jul;17(3):228-236
pubmed: 29994316
Orthopedics. 2010 Nov 02;33(11):797
pubmed: 21053892
Knee. 2016 Jan;23(1):149-51
pubmed: 26765862

Auteurs

Yu Yue (Y)

Department of Electronics, Peking University, Beijing, China.

Qiaochu Gao (Q)

Department of Electronics, Peking University, Beijing, China.

Minwei Zhao (M)

Department of Orthopedics, Peking University Third Hospital, and Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China.

Dou Li (D)

Department of Electronics, Peking University, Beijing, China.

Hua Tian (H)

Department of Orthopedics, Peking University Third Hospital, and Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China.

Classifications MeSH