Машинное обучение и искусственный интеллект в прогнозировании, диагностике и лечении заболеваний грудной аорты (обзор литературы). Часть 2

Аневризма и расслоение грудной аорты, невзирая на относительно невысокую распространенность в сравнении с клапанными пороками и ишемической болезнью сердца, являются потенциально фатальными заболеваниями и представляют собой серьезные проблемы здравоохранения. Показания к хирургическому лечению большинства заболеваний грудной аорты устанавливаются преимущественно на основании максимального диаметра аорты в той или иной зоне. В качестве дополнительных факторов риска ассоциированных с аортой осложнений, влияющих на «ужесточение» показаний и снижение «порогового» значения диаметра аорты, рассматриваются врожденные дисплазии соединительной ткани, аномалии грудной аорты (например, коарктация аорты), семейный анамнез аневризм, расслоений аорты и внезапных смертей. Вместе с тем у определенной доли пациентов с аортопатиями расслоения и разрывы аорты развиваются при нормальном или близком к нормальному диаметру грудной аорты в том или ином отделе. На развитие заболеваний аорты и осложнений влияет множество факторов, и оценка вклада в этиологию и патогенез каждого из них непроста. Машинное обучение и математическое моделирование с использованием искусственного интеллекта — активно развивающееся направление компьютерных наук, которое находит применение и в медицине, в частности, в изучении, диагностике и лечении аневризм и расслоений грудной аорты. В статье рассмотрены современные методы анализа данных, прогнозирования развития аневризм и расслоений грудной аорты, планирования лечения заболеваний грудной аорты и предсказания осложнений с помощью машинного обучения и искусственного интеллекта.

1. Michelena HI, Khanna AD, Mahoney D, et al. Incidence of aortic complications in patients with bicuspid aortic valves. Jama. 2011;306(10):1104–12.

2. Meffert P, Tscheuschler A, Beyersdorf F, et al. Characterization of serum matrix metalloproteinase 2/9 levels in patients with ascending aortic aneurysms. Interactive cardiovascular and thoracic surgery. 2017;24(1):20–6.

3. Branchetti E, Bavaria JE, Grau JB, et al. Circulating soluble receptor for advanced glycation end product identifies patients with bicuspid aortic valve and associated aortopathies. Arteriosclerosis, thrombosis, and vascular biology. 2014;34(10):2349–57.

4. Elefteriades JA. Editorial comment: biomarkers for diagnosis in thoracic aortic disease. Cardiology clinics. 28. Netherlands2010. p. 221–2.

5. Peterss S, Bhandari R, Rizzo JA, et al. The Aortic Root: Natural History After Root-Sparing Ascending Replacement in Nonsyndromic Aneurysmal Patients. The Annals of thoracic surgery. 2017;103(3):828–33.

6. Rajput FA, Zeltser R. Aortic valve replacement. StatPearls. Treasure Island (FL): StatPearls Publishing Copyright © 2021, StatPearls Publishing LLC.; 2021.

7. Sirajuddin S, Wang CK, Al-Abri Q, et al. Early results of a modified biological valved conduit for the Bentall procedure. Journal of cardiac surgery. 2019;34(6):412–8.

8. Amisha, Malik P, Pathania M, Rathaur VK. Overview of artificial intelligence in medicine. Journal of family medicine and primary care. 2019;8(7):2328–31.

9. Chandrashekar A, Handa A, Shivakumar N, et al. A deep learning pipeline to automate high-resolution arterial segmentation with or without intravenous contrast. Annals of surgery. 2022;276(6):e1017lpagee27.

10. Bruse JL, Zuluaga MA, Khushnood A, et al. Detecting clinically meaningful shape clusters in medical image data: metrics analysis for hierarchical clustering applied to healthy and pathological aortic arches. IEEE Trans Biomed Eng. 2017;64(10):2373–83.

11. Alhafez BA, Truong VTT, Ocazionez D, Sohrabi S, Sandhu H, Estrera A, et al. Aortic arch tortuosity, a novel biomarker for thoracic aortic disease, is increased in adults with bicuspid aortic valve. International journal of cardiology. 2019;284:84–9.

12. Condemi F, Campisi S, Viallon M, Relationship between ascending thoracic aortic aneurysms hemodynamics and biomechanical properties. IEEE Trans Biomed Eng. 2019.

13. Della Corte A, Bancone C, Buonocore M, et al. Pattern of ascending aortic dimensions predicts the growth rate of the aorta in patients with bicuspid aortic valve. JACC Cardiovascular imaging. 2013;6(12):1301– 10.

14. Krüger T, Sandoval Boburg R, Lescan M, et al. Aortic elongation in aortic aneurysm and dissection: the Tübingen Aortic Pathoanatomy (TAIPAN) project. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery. 2018;54(1):26–33.

15. Catalano C, Agnese V, Gentile G, et al. Atlasbased evaluation of hemodynamic in ascending thoracic aortic aneurysms. Applied Sciences. 2022;12(1):394.

16. Huang LT, Tsai YS, Liou CF, et al. Automated Stanford classification of aortic dissection using a 2-step hierarchical neural network at computed tomography angiography. European radiology. 2022;32(4):2277–85.

17. Adam C, Fabre D, Mougin J, et al. Pre-surgical and post-surgical aortic aneurysm maximum diameter measurement: full automation by artificial intelligence. European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery. 2021;62(6):869–77.

18. Macruz FBC, Lu C, Strout J, et al. Quantification of the thoracic aorta and detection of aneurysm at CT: development and validation of a fully automatic methodology. Radiology Artificial intelligence. 2022;4(2):e210076.

19. Frasconi P, Baracchi D, Giusti B, et al. Twodimensional aortic size normalcy: a novelty detection approach. Diagnostics (Basel, Switzerland). 2021;11(2).

20. Duff L, Scarsbrook AF, Mackie SL, et al. A methodological framework for AI-assisted diagnosis of active aortitis using radiomic analysis of FDG PET-CT images: Initial analysis. J Nucl Cardiol. 2022;29(6):3315– 31.

21. Bollache E, Guzzardi DG, Sattari S, et al. Aortic valve-mediated wall shear stress is heterogeneous and predicts regional aortic elastic fiber thinning in bicuspid aortic valve-associated aortopathy. The Journal of thoracic and cardiovascular surgery. 2018;156(6):2112– 20.e2.

22. Franco P, Sotelo J, Guala A, et al. Identification of hemodynamic biomarkers for bicuspid aortic valve induced aortic dilation using machine learning. Comput Biol Med. 2022;141:105147.

23. Masri A, Kalahasti V, Svensson LG, et al. Aortic cross-sectional area/height ratio and outcomes in patients with bicuspid aortic valve and a dilated ascending aorta. Circulation Cardiovascular imaging. 2017;10(6):e006249.

24. Wojnarski CM, Roselli EE, Idrees JJ, et al. Machine-learning phenotypic classification of bicuspid aortopathy. The Journal of thoracic and cardiovascular surgery. 2018;155(2):461–9.e4.

25. Wang D, Zhang R, Zhu J, et al. Neural network fusion: a novel CT-MR Aortic Aneurysm image segmentation method. Proceedings of SPIE-the International Society for Optical Engineering. 2018;10574.

26. Chen D, Zhang X, Mei Y, et al. Multi-stage learning for segmentation of aortic dissections using a prior aortic anatomy simplification. Medical image analysis. 2021;69:101931.

27. Cao L, Shi R, Ge Y, et al. Fully automatic segmentation of type B aortic dissection from CTA images enabled by deep learning. European journal of radiology. 2019;121:108713.

28. Liang L, Liu M, Martin C, et al. A machine learning approach to investigate the relationship between shape features and numerically predicted risk of ascending aortic aneurysm. Biomechanics and modeling in mechanobiology. 2017;16(5):1519–33.

29. Fantazzini A, Esposito M, Finotello A, et al. 3D automatic segmentation of aortic computed tomography angiography combining multi-view 2D convolutional neural networks. Cardiovascular engineering and technology. 2020;11(5):576–86.

30. Liu M, Liang L, Ismail Y, et al. Computation of a probabilistic and anisotropic failure metric on the aortic wall using a machine learning-based surrogate model. Comput Biol Med. 2021;137:104794.

31. Chen W, Huang H, Huang J, et al. Deep learningbased medical image segmentation of the aorta using XR-MSF-U-Net. Computer methods and programs in biomedicine. 2022;225:107073.

32. Nita CI, Puiu A, Bunescu D, et al. Personalized pre- and post-operative hemodynamic assessment of aortic coarctation from 3D rotational angiography. Cardiovascular engineering and technology. 2022;13(1):14–40.

33. Talebi S, Madani MH, Madani A, et al. Machine learning for endoleak detection after endovascular aortic repair. Scientific reports. 2020;10(1):18343.

34. Zhou C, Wang R, Jiang W, et al. Machine learning for the prediction of acute kidney injury and paraplegia after thoracoabdominal aortic aneurysm repair. Journal of cardiac surgery. 2020;35(1):89–99.

35. Saitta S, Sturla F, Caimi A, et al. A deep learningbased and fully automated pipeline for thoracic aorta geometric analysis and planning for endovascular repair from computed tomography. Journal of digital imaging. 2022;35(2):226–39.

36. Zhou M, Shi Z, Li X, et al. Prediction of distal aortic enlargement after proximal repair of aortic dissection using machine learning. Annals of vascular surgery. 2021;75:332–40.

37. Dong Y, Que L, Jia Q, et al. Predicting reintervention after thoracic endovascular aortic repair of Stanford type B aortic dissection using machine learning. European radiology. 2022;32(1):355–67.

38. Larentzakis A, Lygeros N. Artificial intelligence (AI) in medicine as a strategic valuable tool. The Pan African medical journal. 2021;38:184.

39. Semenova D, Zabirnyk A, Lobov A, et al. Multiomics of in vitro aortic valve calcification. Frontiers in cardiovascular medicine. 2022;9:1043165.

40. Aguiar M, Trujillo M, Chaves D, et al. mHealth apps using behavior change techniques to self-report data: systematic review. JMIR mHealth and uHealth. 2022;10(9):e33247.

41. Labovitz DL, Shafner L, Reyes Gil M, et al. Using artificial intelligence to reduce the risk of nonadherence in patients on anticoagulation therapy. Stroke. 2017;48(5):1416–-9