Image-based clustering analysis of calcification patterns in bioprosthetic heart valves

Objective: to identify key patterns of calcification in explanted bioprosthetic heart valves (BHVs) using cluster analysis of computed tomography-derived graphical data.

Materials and methods. The study included 11 UniLine BHVs that were routinely explanted during reoperations for structural valve dysfunction. Computed tomography was used to obtain DICOM images of each sample, followed by generation of maximum intensity projections and segmentation of the valves into individual leaflets (n = 33). The images were pre-processed using binary thresholding to differentiate calcified regions from non-calcified biological tissue. Cluster analysis was performed using various algorithms: Gaussian mixture models, Ordering Points To Identify the Clustering Structure (OPTICS), k-means clustering, agglomerative (hierarchical) clustering, and spectral clustering. A basic quantitative method assessing the proportion of pixels corresponding to calcified areas was used for comparison. The performance of clustering algorithms was evaluated using the silhouette score. The presence of calcium deposits in the valves and the accuracy of binary thresholding were further verified histologically by alizarin red S staining of valve cryosections.

Results. Data preprocessing based on image binarization yielded a maximum silhouette score of 0.55. Among the clustering algorithms, the highest silhouette scores were achieved with the agglomerative (0.55) and k-means (0.54) methods; however, both demonstrated substantial data imbalance, with up to 85% of samples grouped within a single cluster, limiting their practical applicability. The most balanced clustering was achieved using spectral clustering (silhouette score 0.45) and the basic quantitative approach (0.44). Both methods identified three distinct patterns of bioprosthetic valve leaflet calcification: (1) non-calcified leaflets, (2) partial calcification, and (3) total calcification.

Conclusion. Three key calcification patterns were identified in explanted BHVs – absence of calcium, partial calcification, and total calcification. Spectral clustering and the basic quantitative method demonstrated the most balanced results, while other algorithms showed pronounced cluster imbalance. Heat map analysis revealed that in partial calcification, mineral deposition typically begins in the commissural and dome regions of the leaflets, near the free edge, and in total calcification, extends across the entire dome and leaflet base.

1. БBokeriya LA, Milievskaya EB, Pryanishnikov VV, Yurlov IA. Serdechno-sosudistaya khirurgiya – 2022. Bolezni i vrozhdennye anomalii sistemy krovoobrashcheniya. М.: NMITs SSKh im. A.N. Bakuleva, 2023; 344. (In Russ.).

2. Alekyan BG, Grigor’yan AM, Staferov AV, Karapetyan NG. Rentgenendovaskulyarnaya diagnostika i lechenie zabolevaniy serdtsa i sosudov v Rossiyskoy Federatsii – 2021 god. Endovaskulyarnaya khirurgiya. 2022; 9: 1–254. (In Russ.). doi: 10.24183/2409-40802022-9S.

3. Wang M, Furnary AP, Li H-F, Grunkemeier GL. Bioprosthetic Aortic Valve Durability: A Meta-Regression of Published Studies. Ann Thorac Surg. 2017 Sep; 104 (3): 1080–1087. doi: 10.1016/j.athoracsur.2017.02.011.

4. Kaneko T, Vassileva CM, Englum B, Kim S, Yammine M, Brennan M et al. Contemporary outcomes of repeat aortic valve replacement: A benchmark for transcatheter valvein-valve procedures. Ann Thorac Surg. 2015 Oct; 100 (4): 1298–1304. doi: 10.1016/j.athoracsur.2015.04.062.

5. Klyshnikov KU, Ovcharenko EA, Stasev AN, Scheglova NA, Odarenko YuN, Khalivopulo IK et al. In vitro study of a biological prosthetic valve for seamless fixation. Russian Journal of Transplantology and Artificial Organs. 2017; 19 (4): 61–69. (In Russ, English abstract). doi: 10.15825/1995-1191-2017-4-61-69.

6. Balsam LB, Grossi EA, Greenhouse DG, Ursomanno P, Deanda A, Ribakove GH et al. Reoperative valve surgery in the elderly: Predictors of risk and long-term survival. Ann Thorac Surg. 2010 Oct; 90 (4): 1195–1200. doi: 10.1016/j.athoracsur.2010.04.057.

7. Marro M, Kossar AP, Xue Y, Frasca A, Levy RJ, Ferrari G. Noncalcific Mechanisms of Bioprosthetic Structural Valve Degeneration. J Am Heart Assoc. 2021 Feb 2; 10 (3): e018921. doi: 10.1161/JAHA.120.018921.

8. Capodanno D, Petronio AS, Prendergast B, Eltchaninoff H, Vahanian A, Modine T et al. Standardized definitions of structural deterioration and valve failure in assessing long-term durability of transcatheter and surgical aortic bioprosthetic valves: a consensus statement from the European Association of Percutaneous Cardiovascular Interventions (EAPCI) endorsed by the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2017 Dec 1; 38 (45): 3382–3390. doi: 10.1093/eurheartj/ehx303.

9. Dvir D, Bourguignon T, Otto CM, Hahn RT, Rosenhek R, Webb JG et al. Standardized Definition of Structural Valve Degeneration for Surgical and Transcatheter Bioprosthetic Aortic Valves. Circulation. 2018 Jan 23; 137 (4): 388–399. doi: 10.1161/CIRCULATIONAHA.117.030729.

10. Kostyunin AE, Yuzhalin AE, Rezvova MA, Ovcharenko EA, Glushkova TV, Kutikhin AG. Degeneration of Bioprosthetic Heart Valves: Update 2020. J Am Heart Assoc. 2020 Oct 20; 9 (19): e018506. doi: 10.1161/JAHA.120.018506.

11. Whelan A, Williams E, Fitzpatrick E, Murphy BP, Gunning PS, O’Reilly D, Lally C. Collagen fibre-mediated mechanical damage increases calcification of bovine pericardium for use in bioprosthetic heart valves. Acta Biomater. 2021 Jul 1; 128: 384–392. doi: 10.1016/j.actbio.2021.04.046.

12. Tsolaki E, Corso P, Zboray R, Avaro J, Appel C, Liebi M et al. Multiscale multimodal characterization and simulation of structural alterations in failed bioprosthetic heart valves. Acta Biomater. 2023 Oct 1; 169: 138–154. doi: 10.1016/j.actbio.2023.07.044.

13. Kim H, Lu J, Sacks MS, Chandran KB. Dynamic Simulation of Bioprosthetic Heart Valves Using a Stress Resultant Shell Model. Ann Biomed Eng. 2008 Feb; 36 (2): 262–275. doi: 10.1007/s10439-007-9409-4.

14. Halevi R, Hamdan A, Marom G, Mega M, Raanani E, Haj-Ali R. Progressive aortic valve calcification: Threedimensional visualization and biomechanical analysis. J Biomech. 2015 Feb 5; 48 (3): 489–497. doi: 10.1016/j.jbiomech.2014.12.004.

15. Thubrikar MJ, Deck JD, Aouad J, Nolan SP. Role of mechanical stress in calcification of aortic bioprosthetic valves. J Thorac Cardiovasc Surg. 1983 Jul; 86 (1): 115–125. doi: 10.1016/S0022-5223(19)39217-7.

16. Morany A, Lavon K, Halevi R, Haj-Ali N, Bluestein D, Raanani E et al. Fragmentation of Different Calcification Growth Patterns in Bicuspid Valves During Balloon Valvuloplasty Procedure. Ann Biomed Eng. 2023 May; 51 (5): 1014–1027. doi: 10.1007/s10439-022-03115-8.

17. Luraghi G, Matas JFR, Beretta M, Chiozzi N, Iannetti L, Migliavacca F. The impact of calcification patterns in transcatheter aortic valve performance: a fluid-structure interaction analysis. Comput Methods Biomech Biomed Engin. 2021 Mar; 24 (4): 375–383. doi: 10.1080/10255842.2020.1817409.

18. Pestiaux C, Pyka G, Quirynen L, De Azevedo D, Vanoverschelde J-L, Lengelé B et al. 3D histopathology of stenotic aortic valve cusps using ex vivo microfocus computed tomography. Front Cardiovasc Med. 2023 Apr 25; 10: 1129990. doi: 10.3389/fcvm.2023.1129990.

19. Cartlidge TRG, Doris MK, Sellers SL, Pawade TA, White AC, Pessotto R et al. Detection and Prediction of Bioprosthetic Aortic Valve Degeneration. J Am Coll Cardiol. 2019 Mar 19; 73 (10): 1107–1119. doi: 10.1016/j.jacc.2018.12.056.

20. Bishop C. Pattern Recognition and Machine Learning. Journal of Electronic Imaging. 2006 Jan; 16 (4): 140– 155. doi: 10.1117/1.2819119.

21. Ankerst M, Breunig MM, Kriegel H-P, Sander J. OPTICS: ordering points to identify the clustering structure. ACM SIGMOD Rec. 1999 Jun; 28 (2): 49–60. doi: 10.1145/304181.304187.

22. Murtagh F, Contreras P. Algorithms for hierarchical clustering: an overview. WIREs Data Min Knowl Discov. 2012 Dec 7; 2 (1): 86–97. doi: 10.1002/widm.53.

23. Rousseeuw PJ. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. J Comput Appl Math. 1987 Nov; 20: 53–65. doi: 10.1016/0377-0427(87)90125-7.

24. Van der Maaten L, Hinton G. Visualizing High-Dimensional Data Using t-SNE. J Mach Learn Res. 2008; 9 (86): 2579–2605.

25. Leong SW, Soor GS, Butany J, Henry J, Thangaroopan M, Leask RL. Morphological findings in 192 surgically excised native mitral valves. Can J Cardiol. 2006 Oct; 22 (12):1055–1061.doi:10.1016/S0828-282X(06)70321-X.

26. Rutkovskaya NV, Kagan ES, Kondyukova NV, Kuzmina OK, Barbarash LS. Prediction of Risk of Calcification of Heart Valve Bioprostheses on the Basis of Comprehensive Assessment of Recipients Clinical Factors and Their Compliance With Therapy. Kardiologiia. 2018; 58 (10): 27–33. (In Russ, English abstract). doi: 10.18087/cardio.2018.10.10182.

27. Mahjoub H, Mathieu P, Larose E, Dahou A, Sénéchal M, Dumesnil J-G et al. Determinants of aortic bioprosthetic valve calcification assessed by multidetector CT. Heart. 2015 Mar; 101 (6): 472–477. doi: 10.1136/heartjnl-2014-306445.

28. Frasca A, Xue Y, Kossar AP, Keeney S, Rock C, Zakharchenko A et al. Glycation and Serum Albumin Infiltration Contribute to the Structural Degeneration of Bioprosthetic Heart Valves. JACC Basic to Transl Sci. 2020 Aug 5; 5 (8): 755–766. doi: 10.1016/j.jacbts.2020.06.008.

29. Sakaue T, Koyama T, Nakamura Y, Okamoto K, Kawashima T, Umeno T et al. Bioprosthetic Valve Deterioration. JACC Basic to Transl Sci. 2023 May 10; 8 (7): 862–880. doi: 10.1016/j.jacbts.2023.01.003.

30. Ovcharenko EA, Klyshnikov KU, Glushkova ТV, Batranin АV, Rezvova МА, Kudryavtseva YА, Barbarash LS. Evaluation of a failed heart valve bioprosthesis using microcomputed tomography. Sovremennye tehnologii v medicine. 2017; 9 (3): 15–22. (In Russ.). https://doi.org/10.17691/stm2017.9.3.02.