Artificial intelligence for the classification of fractures around the knee in adults according to the 2018 AO/OTA classification system.

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
Read Online Read More Add to Saved list
  • Additional Information
    • Source:
      Publisher: Public Library of Science Country of Publication: United States NLM ID: 101285081 Publication Model: eCollection Cited Medium: Internet ISSN: 1932-6203 (Electronic) Linking ISSN: 19326203 NLM ISO Abbreviation: PLoS One Subsets: MEDLINE
    • Publication Information:
      Original Publication: San Francisco, CA : Public Library of Science
    • Subject Terms:
      Artificial Intelligence*; Femoral Fractures/*diagnostic imaging ; Image Processing, Computer-Assisted/*methods ; Tibial Fractures/*diagnostic imaging; Humans
    • Abstract:
      Background: Fractures around the knee joint are inherently complex in terms of treatment; complication rates are high, and they are difficult to diagnose on a plain radiograph. An automated way of classifying radiographic images could improve diagnostic accuracy and would enable production of uniformly classified records of fractures to be used in researching treatment strategies for different fracture types. Recently deep learning, a form of artificial intelligence (AI), has shown promising results for interpreting radiographs. In this study, we aim to evaluate how well an AI can classify knee fractures according to the detailed 2018 AO-OTA fracture classification system.
      Methods: We selected 6003 radiograph exams taken at Danderyd University Hospital between the years 2002-2016, and manually categorized them according to the AO/OTA classification system and by custom classifiers. We then trained a ResNet-based neural network on this data. We evaluated the performance against a test set of 600 exams. Two senior orthopedic surgeons had reviewed these exams independently where we settled exams with disagreement through a consensus session.
      Results: We captured a total of 49 nested fracture classes. Weighted mean AUC was 0.87 for proximal tibia fractures, 0.89 for patella fractures and 0.89 for distal femur fractures. Almost ¾ of AUC estimates were above 0.8, out of which more than half reached an AUC of 0.9 or above indicating excellent performance.
      Conclusion: Our study shows that neural networks can be used not only for fracture identification but also for more detailed classification of fractures around the knee joint.
      Competing Interests: The authors have read the journal’s policy and the authors of this manuscript have the following competing interests: MG, OS, and AS are co-founders and shareholders in DeepMed AB. While this puts us at a financial competing interest, the company has limited activity with currently no revenue stream, no external investors and no pending patents. There are no patents, products in development or marketed products to declare. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
    • References:
      Nature. 2017 Feb 2;542(7639):115-118. (PMID: 28117445)
      Acta Orthop. 2019 Aug;90(4):394-400. (PMID: 30942136)
      Onco Targets Ther. 2015 Aug 04;8:2015-22. (PMID: 26346558)
      Acta Orthop. 2018 Aug;89(4):468-473. (PMID: 29577791)
      AJR Am J Roentgenol. 2017 Apr;208(4):739-749. (PMID: 28026210)
      Acta Orthop. 2017 Dec;88(6):581-586. (PMID: 28681679)
      Proc Natl Acad Sci U S A. 2018 Nov 6;115(45):11591-11596. (PMID: 30348771)
      Acta Orthop. 2021 Feb;92(1):102-108. (PMID: 33103536)
      Caspian J Intern Med. 2013 Spring;4(2):627-35. (PMID: 24009950)
      Clin Radiol. 2018 May;73(5):439-445. (PMID: 29269036)
      IEEE Trans Neural Netw Learn Syst. 2014 May;25(5):845-69. (PMID: 24808033)
      Comput Methods Programs Biomed. 2019 Apr;171:27-37. (PMID: 30902248)
      J Thorac Oncol. 2010 Sep;5(9):1315-6. (PMID: 20736804)
      JAMA. 2017 Dec 12;318(22):2211-2223. (PMID: 29234807)
      Emerg Radiol. 2011 Aug;18(4):279-83. (PMID: 21394519)
      Biochem Med (Zagreb). 2012;22(3):276-82. (PMID: 23092060)
      BMC Emerg Med. 2006 Feb 16;6:4. (PMID: 16483365)
      Med Image Anal. 2017 Jan;35:303-312. (PMID: 27497072)
      Radiographics. 2017 Mar-Apr;37(2):505-515. (PMID: 28212054)
      Skeletal Radiol. 2019 Feb;48(2):239-244. (PMID: 29955910)
      World J Orthop. 2016 Sep 18;7(9):584-91. (PMID: 27672572)
      J Orthop Trauma. 2018 Jan;32 Suppl 1:S1-S170. (PMID: 29256945)
      Shanghai Arch Psychiatry. 2018 Jun 25;30(3):207-212. (PMID: 30858674)
      Clin Orthop Relat Res. 2019 Nov;477(11):2482-2491. (PMID: 31283727)
      J Bone Joint Surg Am. 2014 Jan 15;96(2):144-50. (PMID: 24430414)
      Science. 2006 Jul 28;313(5786):504-7. (PMID: 16873662)
    • Publication Date:
      Date Created: 20210401 Date Completed: 20210915 Latest Revision: 20210915
    • Publication Date:
      20221213
    • Accession Number:
      PMC8016258
    • Accession Number:
      10.1371/journal.pone.0248809
    • Accession Number:
      33793601