Ortopedi ve Travmatolojide Yapay Zeka Uygulamaları

Abstract

Ortopedide yapay zeka konusunda yayınlanan çalışmalara bakıldığında röntgen grafisinden kırık tanıma ve eklem dejenerasyon sınıflandırma sistemleri oldukça popülerdir. Bu durumun başlıca nedeni ortopedistlerin çoğu zaman bu hastalıkların tanısında sadece röntgen grafisine bakarak karar vermeleridir. Ayrıca röntgen grafisinin manyetik rezonans görüntüleme ve bilgisayarlı tomografi görüntülerine kıyasla daha kolay işlenmesidir. Yapılan çalışmalar geliştirilen algoritmaların verdiği kararlarda ortopedist olmayan hekimlerden daha iyi ve neredeyse bir ortopedik cerraha yakınlıkta tanı koyma potansiyeli göstermektedir. Yapay zeka uygulamaları sadece görüntü odaklı tanı koyma alanında değil, preop değerlendirmede, postop takip süreçlerinin yönetiminde, hasta memnuniyetinin artırılmasında ve maliyet etkin kişisel çözümler sunarak kaynakların daha verimli kullanılmasında gelecek vaad etmektedir. Yakın gelecekte doğal dil işleme tekniklerinin de etkin hale gelmesiyle hastayı dinleyen, raporları yorumlayan sistemlerin de hayatımıza girmesi olasıdır.

Keywords

• Yapay zeka
• Doğal dil işleme
• Ortopedi ve travmatoloji
• Karar destek sistemi

Artificial Intelligence Applications In Orthopaedics & Traumatology

Abstract

The published studies about artificial intelligence in orthopedics show that fracture recognition and joint degeneration classification systems from X-ray are very popular. The main reason for this situation is that orthopedists often decide to only look at the x-ray in the diagnosis of these diseases. In addition, the easier processing of X-ray radiography compared to magnetic resonance imaging and computed tomography images increases the interest in these issues. Studies show the potential of making a diagnosis beter than non-orthopedic physicians and almost closer to an orthopedic surgeon in the developed algorithms’ decisions. Artificial intelligence applications are promising in image-oriented diagnosis, in preoperative evaluation, management of postoperative follow-up processes, increasing patient satisfaction and providing cost-effective personal solutions, making the use of resources more efficient. In the near future, with the use of natural language processing techniques systems that listen to the patient and interpret the reports may also enter our lives.

Keywords

• Artificial intelligence
• Natural language processing
• Orthopedics and traumatology
• Decision support system

References

1. 1- Berg, Hans E. WillIntelligent Machine Learning RevolutionizeOrthopedicImaging? ActaOrthopaedica. 2017;88(6), 577–577. DOI: 10.1080/17453674.2017.1387732.
2. 2- Olczak, J.,Fahlberg N., Maki A, et al. ArtificialIntelligenceforAnalyzingOrthopedicTraumaRadiographs. ActaOrthopaedica. 2017;88(6):581-586. DOI: 10.1080/17453674.2017.1344459
3. 3-Audigé, Laurent, et al. How ReliableAreReliabilityStudies of FractureClassifications? A SystematicReview of TheirMethodologies. ActaOrthopaedicaScandinavica. 2004;75(2) 184–194. DOI: 10.1080/00016470412331294445.
4. 4- Jeana,Z.,Elainea, D., Danielc, PA. et.al. Osteoporosisepidemiologyusinginternationalcohorts, CurrentOpinion in Rheumatology: 2020;32(4)-387-393 DOI: 10.1097/BOR.0000000000000722
5. 5- Beyaz, S.,Acici K., Sumer E. FemoralNeckFractureDetection in X-Ray Images Using Deep Learning and GeneticAlgorithmApproaches. JointDiseases and RelatedSurgery. 2020;31(2)175–183. DOI: 10.5606/ehc.2020.72163.
6. 6-Norman B.,Pedoia V., Noworolski A., et al. Applyingdenselyconnectedconvolutionalneuralnetworksforstagingosteoarthritisseverityfromplainradiographs. J DigitImaging 2019;32(03):471– 477. DOI: 10.1007/s10278-018-0098-3
7. 7-Lindsey R.,Daluiski A., Chopra S., et al. DeepNeural Network ImprovesFractureDetectionbyClinicians. Proceedings of the National Academy of Sciences. 2018, vol. 115, no. 45, 11591– 11596. DOI: 10.1073/pnas.1806905115
8. 8-Urakawa T.,Tanaka Y., Goto S., et al. DetectingIntertrochantericHipFractureswithOrthopedist-Level Accuracy Using a DeepConvolutionalNeural Network. SkeletalRadiology, 2019 Feb;48(2):239-244.DOI:10.1007/s00256-018-3016-3.

9. 9-Zdolsek G, Chen Y., Bögl HP., et al. Deepneuralnetworkswithpromisingdiagnosticaccuracyfor the classificationof atypicalfemoralfractures, ActaOrthopaedica 2021, DOI: 10.1080/17453674.2021.1891512
10. 10- Cabitza, F.,Locoro, A., Banfi, G. Machine learning in orthopedics: A literaturereview. Front BioengBiotechnol. 2018 Jun27;6:75. DOI: 10.3389/fbioe.2018.00075
11. 11-Bovonratwet P, Shen TS, Islam W, Ast MP, Haas SB, Su EP. Natural Language Processing of Patient-ExperienceCommentsAfterPrimary Total KneeArthroplasty. J Arthroplasty. 2021;36(3):927-934. DOI:10.1016/j.arth.2020.09.055
12. 12-Thirukumaran CP, Zaman A, Rubery PT, et al. Natural Lan - guage Processingfor the Identification of Surgical Site Infections in Orthopaedics. J Bone JointSurgAm. 2019;101(24):2167-2174. DOI:10.2106/JBJS.19.00661
13. 13-Wyles CC, Tibbo ME, Fu S, et al. Use of Natural Language ProcessingAlgorithmsToIdentifyCommon Data Elements in OperativeNotesfor Total HipArthroplastyJ Bone JointSurgAm. 2019;101(21):1931-1938. DOI:10.2106/JBJS.19.0007
14. 14-Kolanu N, Brown AS, Beech A, Center JR, White CP. Nat - ural languageprocessing of radiologyreportsfor the identification of patientswithfracture. ArchOsteoporos. 2021 Jan 6;16(1):6. doi: 10.1007/s11657-020-00859-5.
15. 15-Schwartz WB. Medicine and the computer: the promise and problems of change. N Engl J Med. 1970;283(23):1257–1264. doi:10.1056/NEJM197012032832305
16. 16-Loftus TJ, Tighe PJ, Filiberto AC, et al. ArtificialIntelligence and SurgicalDecision-making. JAMA Surg. 2020;155(2):148- 158. doi:10.1001/jamasurg.2019.4917
17. 17-Stacey D, Légaré F, Lewis K, et al. Decisionaidsforpeople - facinghealthtreatmentorscreeningdecisions. Cochrane Database SystRev. 2017;4: CD001431. doi:10.1002/14651858.CD001431.pub5
18. 18- Kim JS.,Merril RK., Arvind V., et al. Examining the Ability of ArtificialNeural Networks Machine Learning ModelstoAccu - ratelyPredictComplicationsFollowingPosteriorLumbarSpine - Fusion. Spine, 2018 Jun 15;43(12):853-860. DOI: 10.1097/ BRS.0000000000002442.
19. 19-Pereira NRP, Janssen, SJ.,Dijk, EV., et al. Development of a prognosticsurvivalalgorithmforpatientswithmetastaticspinedis - ease. J Bone JointSurgAm, 2016 Nov 2;98(21):1767-1776. DOI: 10.2106/JBJS.15.00975
20. 20-Piccioli, Andrea, et al. How Do WeEstimateSurvival? Exter - nalValidation of a ToolforSurvivalEstimation in Patientswith - Metastatic Bone Disease—Decision Analysis and Comparison of Three International PatientPopulations. BMC Cancer. 2015 May 22;15:424.DOI: 10.1186/s12885-015-1396-5
21. 21- Forsberg, J. A.,Wedin, R., Boland, P. J.et.al Can weesti - mateshort- and intermediate-termsurvival in patientsundergo - ingsurgeryformetastatic bone disease? ClinOrthopRelatRes. 2017;475(4):1252-1261.DOI: 10.1007/s11999-016-5187-3
22. 22-Fontana, MA.,Lyman S., Sarker GK., et al.Can Machine Learning AlgorithmsPredictWhichPatientsWillAchieveMinimal - lyClinicallyImportantDifferencesFrom Total JointArthroplasty? ClinOrthopRelatRes. 2019 Jun;477(6):1267-1279. DOI: 10.1097/ CORR.0000000000000687
23. 23-Cilla, Myriam, et al. Machine Learning Techniquesfor the Optimization of JointReplacements: Application to a Short-Stem - HipImplant. PLoSOne. 2017 Sep 5;12(9):e0183755.DOI: 10.1371/journal.pone.0183755.
24. 24-Oosterhoff JHF, Doornberg JN; Machine Learning Consorti - um. Artificialintelligence in orthopaedics: falsehopeor not? A nar - rativereviewalong the line of Gartner’shypecycle. EFORT Open Rev. 2020 Oct 26;5(10):593-603. DOI: 10.1302/2058-5241.5.190092