Prediction of impaired consolidation of limb long-bone fractures using neural network analysis

Introduction Impaired reparative regeneration in patients with fractures is the most common complication; immunogenetic mechanisms play a leading role in its pathogenesis. Many researchers are engaged in the search for an "ideal" diagnostic marker. For this purpose, neural networks have been increasingly used, which allow not only to predict various pathological conditions but also to determine reliable options for prevention and treatment.

The purpose of the study was to evaluate the effectiveness of predicting impaired consolidation of long-bone fractures of the extremities using the neural network data analysis.

Material and methods We examined 108 young patients (WHO classification) with fractures of lower limb long bones. The clinical comparison group consisted of 62 patients without complications at the age of 34.5 [18; 44] years. The study group included 46 patients of similar age (36 [18; 44]) years and gender with delayed consolidation. The control group included 92 practically healthy individuals. Exclusion criteria from the study were any concomitant disease, other location and nature of injuries, alcoholism, as well as inaccurate reduction of bone fragments, and repeated operations. Patients who received antiresorption therapy and calcium supplements in the prehospital stage were also excluded. Laboratory (genetic) studies included determination of carriage of polymorphic molecules — TNFRSF11B-1181(G>C), IL6-174(C>G), TGFβ1-25(Arg>Pro), EGFR-2073(A>T) and VDR(BsmI283G>A). Amplification was carried out using primer sets Litekh-SNP (Russia). The risk of developing delayed consolidation was assessed using SPSS Statistics Version 25.0 (Neural Networks module). The predictive performance of the neural network was assessed using ROC analysis.

Results For determining the importance of the independent variable, the following gradation was noted: TGFβ1-25(Arg>Pro) gene polymorphism — 100 %; gene polymorphism TNFRSF11B-1181(G>C) — 97.1 %, gene polymorphism VDR-BsmI283(G>A) — 34.7 %; IL6-174(C>G) gene polymorphism — 31.5 %; polymorphism of the EGFR-2073(A>T) gene — 15.3 %. The percentage of incorrect predictions was 8.3 %. Area under the curve of ROC analysis (AUC) = 0.91[0.85–0.98], p < 0.001. The specificity of the resulting model is 0.95 %, sensitivity is 0.87 %, accuracy is 91.7 %.

Conclusion The use of the neural network for predicting delayed consolidation of fractures using data on the carriage of certain gene polymorphisms has a sufficient degree of accuracy (91.7 %), which indicates that the introduction of the neural network analysis into practical medicine is promising.

1. Бондаренко А.В., Гусейнов Р.Г., Герасимова О.А. и др. Частота, факторы риска, особенности диафизарных несращений длинных костей нижних конечностей. Политравма. 2023;(2):36-44. doi: 10.24412/1819-1495-2023-2-36-44.

2. Федоров В.Г., Кузин И.В. Результаты лечения переломов диафиза бедренной кости блокируемым интрамедуллярным и накостным остеосинтезом (итоги за 10 лет). Acta biomedica scientifica. 2023;8(5):166-173. doi: 10.29413/ABS.2023-8.5.18.

3. Wildemann B, Ignatius A, Leung F, et al. Non-union bone fractures. Nat Rev Dis Primers. 2021;7(1):57. doi: 10.1038/s41572-021-00289-8.

4. Мироманов А.М., Гусев К.А., Старосельников А.Н. и др. Современные генетические и иммунологические аспекты патогенеза нарушения консолидации переломов (обзор литературы). Acta Biomedica Scientifica. 2022;7(2):49-64. doi: 10.29413/ABS.2022-7.2.6.

5. Afrasiabian B, Eftekhari М. Prediction of mode I fracture toughness of rock using linear multiple regression and gene expression programming. J Rock Mech Geotech Eng. 2022;14(5):1421-1432. doi: 10.1016/j.jrmge.2022.03.008.

6. Cui Y, Hu X, Zhang C, Wang K. The genetic polymorphisms of key genes in WNT pathway (LRP5 and AXIN1) was associated with osteoporosis susceptibility in Chinese Han population. Endocrine. 2022;75(2):560-574. doi: 10.1007/s12020-021-02866-z.

7. Ding ZC, Lin YK, Gan YK, Tang TT. Molecular pathogenesis of fracture nonunion. J Orthop Translat. 2018;14:45-56. doi: 10.1016/j.jot.2018.05.002.

8. MacEachern SJ, Forkert ND. Machine learning for precision medicine. Genome. 2021;64(4):416-425. doi: 10.1139/gen-2020-0131.

9. Pasini A. Artificial neural networks for small dataset analysis. J Thorac Dis. 2015;7(5):953-960. doi: 10.3978/j.issn.2072-1439.2015.04.61.

10. Curchoe CL, Bormann CL. Artificial intelligence and machine learning for human reproduction and embryology presented at ASRM and ESHRE 2018. J Assist Reprod Genet. 2019;36(4):591-600. doi: 10.1007/s10815-019-01408-x.

11. Plumarom Y, Wilkinson BG, Willey MC, et al. Sensitivity and specificity of modified RUST score using clinical and radiographic findings as a gold standard. Bone Jt Open. 2021;2(10):796-805. doi: 10.1302/2633-1462.210.BJO-2021-0071.R1.

12. Lang TA, Altman DG. Statistical analyses and methods in the published literature: The SAMPL guidelines. Medical Writing. 2016;25(3):31-36. doi: 10.18243/eon/2016.9.7.4.

13. Мудров В.А. Алгоритм применения ROC-анализа в биомедицинских исследованиях с помощью пакета программ SPSS. Забайкальский медицинский вестник. 2021;(1):148-153. doi: 10.52485/19986173_2021_1_148.

14. Симбирцев А.С. Иммунофармакологические аспекты системы цитокинов. Бюллетень сибирской медицины. 2019;18(1):84-95. doi: 10.20538/1682-0363-2019-1-84-95.

15. Schaettler MO, Richters MM, Wang AZ, et al. Characterization of the Genomic and Immunologic Diversity of Malignant Brain Tumors through Multisector Analysis. Cancer Discov. 2022;12(1):154-171. doi: 10.1158/2159-8290.CD-21-0291.

16. Akbarov AN, Ziyadullayeva NS, Reimnazarova GD, Nurullaeva MU. Peculiarities of osteoreparation in case of bone defect replacement with bioactive glass in combination with an antibiotic. British View. 2022;7(1):101-105. doi: 10.5281/zenodo.6571286.

17. Ding ZC, Lin YK, Gan YK, Tang TT. Molecular pathogenesis of fracture nonunion. J Orthop Translat. 2018;14:45-56. doi: 10.1016/j.jot.2018.05.002.

18. Zimmermann G, Schmeckenbecher KHK, Boeuf S, et al. Differential gene expression analysis in fracture callus of patients with regular and failed bone healing. Injury. 2012;43(3):347-356. doi: 10.1016/j.injury.2011.10.031.

19. Waki T, Lee SY, Niikura T, et al. Profiling microRNA expression in fracture nonunions: Potential role of microRNAs in nonunion formation studied in a rat model. Bone Joint J. 2015;97-B(8):1144-1451. doi: 10.1302/0301-620X.97B8.34966.

20. Wang H, Xie Z, Hou T, et al. MiR-125b Regulates the Osteogenic Differentiation of Human Mesenchymal Stem Cells by Targeting BMPR1b. Cell Physiol Biochem. 2017;41(2):530-542. doi: 10.1159/000457013.

21. He B, Zhang ZK, Liu J, et al. Bioinformatics and microarray analysis of mirnas in aged female mice model Implied new molecular mechanisms for impaired fracture healing. Int J Mol Sci. 2016;17(8):1260. doi: 10.3390/ijms17081260.

22. Guimarães JM, Guimarães IC, Duarte ME, et al. Polymorphisms in BMP4 and FGFR1 genes are associated with fracture non-union. J Orthop Res. 2013;31(12):1971-1979. doi: 10.1002/jor.22455.

23. Комков А.А., Мазаев В.П., Рязанова С.В. и др. Основные направления развития искусственного интеллекта в медицине. Научное обозрение. Медицинские науки. 2020;(5):33-40. doi: 10.17513/srms.1141.

24. Bernard de Villeneuve F, Jacquet C, El Kadim B, et al. An artificial intelligence based on a convolutional neural network allows a precise analysis of the alignment of the lower limb. Int Orthop. 2023;47(2):511-518. doi: 10.1007/s00264-022-05634-4.

25. Memiş A, Varlı S, Bilgili F. Semantic segmentation of the multiform proximal femur and femoral head bones with the deep convolutional neural networks in low quality MRI sections acquired in different MRI protocols. Comput Med Imaging Graph. 2020;81:101715. doi: 10.1016/j.compmedimag.2020.101715.

26. Lee SI, Celik S, Logsdon BA, et al. A machine learning approach to integrate big data for precision medicine in acute myeloid leukemia. Nat Commun. 2018;9(1):42. doi: 10.1038/s41467-017-02465-5.

27. Зиганшин А.М., Дикке Г.Б., Мудров В.А. Прогнозирование клинически узкого таза с помощью нейросетевого анализа данных. Акушерство, Гинекология и Репродукция. 2023;17(2):211-220. doi: 10.17749/2313-7347/ob.gyn.rep.2023.382.