Development and validation of an artificial intelligence model for the classification of hip fractures using the AO-OTA framework
DOI:
https://doi.org/10.2340/17453674.2024.40949Keywords:
AO-OTA fracture classification, artificial intelligence, convolutional neural network, deep learning, hip fractureAbstract
Background and purpose: Artificial intelligence (AI) has the potential to aid in the accurate diagnosis of hip fractures and reduce the workload of clinicians. We primarily aimed to develop and validate a convolutional neural network (CNN) for the automated classification of hip fractures based on the 2018 AO-OTA classification system. The secondary aim was to incorporate the model’s assessment of additional radiographic findings that often accompany such injuries.
Methods: 6,361 plain radiographs of the hip taken between 2002 and 2016 at Danderyd University Hospital were used to train the CNN. A separate set of 343 radiographs representing 324 unique patients was used to test the performance of the network. Performance was evaluated using area under the curve (AUC), sensitivity, specificity, and Youden’s index.
Results: The CNN demonstrated high performance in identifying and classifying hip fracture, with AUCs ranging from 0.76 to 0.99 for different fracture categories. The AUC for hip fractures ranged from 0.86 to 0.99, for distal femur fractures from 0.76 to 0.99, and for pelvic fractures from 0.91 to 0.94. For 29 of 39 fracture categories, the AUC was ≥ 0.95.
Conclusion: We found that AI has the potential for accurate and automated classification of hip fractures based on the AO-OTA classification system. Further training and modification of the CNN may enable its use in clinical settings.
Downloads
References
Papadimitriou N, Tsilidis K K, Orfanos P, Benetou V, Ntzani E E, Soerjomataram I, et al. Burden of hip fracture using disability-adjusted life-years: a pooled analysis of prospective cohorts in the CHANCES consortium. Lancet Public Health 2017; 2(5): e239-e246. doi: 10.1016/S2468-2667(17)30046-4. DOI: https://doi.org/10.1016/S2468-2667(17)30046-4
Lu Y, Uppal H S. Hip fractures: relevant anatomy, classification, and biomechanics of fracture and fixation. Geriatr Orthop Surg Rehabil 2019; 10: 2151459319859139. doi: 10.1177/2151459319859139. DOI: https://doi.org/10.1177/2151459319859139
Ward R J, Weissman B N, Kransdorf M J, Adler R, Appel M, Bancroft L W, et al. ACR appropriateness criteria acute hip pain—suspected fracture. J Am Coll Radiol 2014; 11(2): 114-20. doi: 10.1016/j.jacr.2013.10.023. DOI: https://doi.org/10.1016/j.jacr.2013.10.023
Ganeshan D, Rosenkrantz A B, Bassett R L, Williams L, Lenchik L, Yang W. Burnout in academic radiologists in the United States. Acad Radiol 2020; 27(9): 1274-81. doi: 10.1016/j.acra.2019.12.029. DOI: https://doi.org/10.1016/j.acra.2019.12.029
Giger M L. Machine learning in medical imaging. J Am Coll Radiol 2018; 15(3): 512-20. doi: 10.1016/j.jacr.2017.12.028. DOI: https://doi.org/10.1016/j.jacr.2017.12.028
Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks. In: Pereira F, Burges CJ, Bottou L, Weinberger KQ, editors. Adv Neural Inf Process Syst [Internet]. Curran Associates, Inc.; 2012. Available from: https://proceedings.neurips.cc/paper_files/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf
London A J. Artificial intelligence and black-box medical decisions: accuracy versus explainability. Hastings Cent Rep 2019; 49(1): 15-21. doi: 10.1002/hast.973. DOI: https://doi.org/10.1002/hast.973
Fogel A L, Kvedar J C. Artificial intelligence powers digital medicine. NPJ Digit Med 2018; 1(1): 5. doi: 10.1038/s41746-017-0012-2. DOI: https://doi.org/10.1038/s41746-017-0012-2
Olczak J, Fahlberg N, Maki A, Razavian A S, Jilert A, Stark A, et al. Artificial intelligence for analyzing orthopedic trauma radiographs. Acta Orthop 2017 Dec; 88(6): 581-6. doi: 10.1080/17453674.2017.1344459. DOI: https://doi.org/10.1080/17453674.2017.1344459
Anderson M, Anderson S L. How should AI be developed, validated, and implemented in patient care? AMA J Ethics 2019; 21(2): 125-30. doi: 10.1001/amajethics.2019.125. DOI: https://doi.org/10.1001/amajethics.2019.125
Krogue J D, Cheng K V, Hwang K M, Toogood P, Meinberg E G, Geiger E J, et al. Automatic hip fracture identification and functional subclassification with deep learning. Radiol Artif Intell 2020 ; 2(2): e190023. doi: 10.1148/ryai.2020190023. DOI: https://doi.org/10.1148/ryai.2020190023
Cheng C T, Chen C C, Cheng F J, Chen H W, Su Y S, Yeh C N, et al. A human–algorithm integration system for hip fracture detection on plain radiography: system development and validation study. JMIR Med Inform 2020; 8(11): e19416. doi: 10.2196/19416. DOI: https://doi.org/10.2196/19416
Urakawa T, Tanaka Y, Goto S, Matsuzawa H, Watanabe K, Endo N. Detecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network. Skeletal Radiol 2019; 48(2): 239-44. doi: 10.1007/s00256-018-3016-3. DOI: https://doi.org/10.1007/s00256-018-3016-3
Mutasa S, Varada S, Goel A, Wong T T, Rasiej M J. Advanced deep learning techniques applied to automated femoral neck fracture detection and classification. J Digit Imaging 2020; 33(5): 1209-17. doi: 10.1007/s10278-020-00364-8. DOI: https://doi.org/10.1007/s10278-020-00364-8
Meinberg E, Agel J, Roberts C, Karam M, Kellam J. Fracture and dislocation classification compendium—2018. J Orthop Trauma 2018; 32(1): S1-S10. doi: 10.1097/BOT.0000000000001063. DOI: https://doi.org/10.1097/BOT.0000000000001063
Collins G S, Reitsma J B, Altman D G, Moons K G M. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ 2015; 350:g7594. doi: 10.1136/bmj.g7594. DOI: https://doi.org/10.1136/bmj.g7594
National Clinical Guideline Centre (UK). The Management of Hip Fracture in Adults [Internet]. London: Royal College of Physicians (UK); 2011. Available from: http://www.ncbi.nlm.nih.gov/books/NBK83014/: (accessed 2024-01-12).
Möller M, Wolf O, Bergdahl C, Mukka S, Rydberg E M, Hailer N P, et al. The Swedish Fracture Register: ten years of experience and 600,000 fractures collected in a National Quality Register. BMC Musculoskelet Disord 2022; 23(1): 141. doi: 10.1186/s12891-022-05062-w. DOI: https://doi.org/10.1186/s12891-022-05062-w
Palm H, Krasheninnikoff M, Holck K, Lemser T, Foss N B, Jacobsen S, et al. A new algorithm for hip fracture surgery: reoperation rate reduced from 18% to 12% in 2,000 consecutive patients followed for 1 year. Acta Orthop 2012; 83(1): 26-30. doi: 10.3109/17453674.2011.652887. DOI: https://doi.org/10.3109/17453674.2011.652887
Mandrekar J N. Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol 2010; 5(9): 1315-16. doi: 10.1097/JTO.0b013e3181ec173d. DOI: https://doi.org/10.1097/JTO.0b013e3181ec173d
Sato Y, Takegami Y, Asamoto T, Ono Y, Hidetoshi T, Goto R, et al. Artificial intelligence improves the accuracy of residents in the diagnosis of hip fractures: a multicenter study. BMC Musculoskelet Disord 2021; 22(1): 407. doi: 10.1186/s12891-021-04260-2. DOI: https://doi.org/10.1186/s12891-021-04260-2
Langerhuizen D W G, Janssen S J, Mallee W H, Bekerom M P J, Ring D, Kerkhoffs G M M J, et al. What are the applications and limitations of artificial intelligence for fracture detection and classification in orthopaedic trauma imaging? A systematic review. Clin Orthop Rel Res 2019; 477(11): 2482-91. doi: 10.1097/CORR.0000000000000848. DOI: https://doi.org/10.1097/CORR.0000000000000848
Gerriets V, Anderson J, Nappe T M. Acetaminophen. StatPearls [Internet]. Treasure Island, FL: StatPearls Publishing; 2022. Available from: http://www.ncbi.nlm.nih.gov/books/NBK482369/: (accessed 2022-01-13).
Additional Files
Published
How to Cite
License
Copyright (c) 2024 Ehsan Akbarian, Mehrgan Mohammadi, Emilia Tiala, Oscar Ljungberg, Ali Sharif Razavian, Martin Magnéli, Max Gordon
This work is licensed under a Creative Commons Attribution 4.0 International License.
