×

Geometric uncertainty in patient-specific cardiovascular modeling with convolutional dropout networks. (English) Zbl 1507.76263

Summary: We propose a novel approach to generate samples from the conditional distribution of patient-specific cardiovascular models given a clinically acquired image volume. A convolutional neural network architecture with dropout layers is first trained for vessel lumen segmentation using a regression approach, to enable Bayesian estimation of vessel lumen surfaces. This network is then integrated into a path-planning patient-specific modeling pipeline to generate families of cardiovascular models. We demonstrate our approach by quantifying the effect of geometric uncertainty on the hemodynamics for three patient-specific anatomies, an aorto-iliac bifurcation, an abdominal aortic aneurysm and a sub-model of the left coronary arteries. A key innovation introduced in the proposed approach is the ability to learn geometric uncertainty directly from training data. The results show how geometric uncertainty produces coefficients of variation comparable to or larger than other sources of uncertainty for wall shear stress and velocity magnitude, but has limited impact on pressure. Specifically, this is true for anatomies characterized by small vessel sizes, and for local vessel lesions seen infrequently during network training.

MSC:

76Z05 Physiological flows
92B20 Neural networks for/in biological studies, artificial life and related topics
65C05 Monte Carlo methods
92C35 Physiological flow

Software:

TetGen; Adam; SimVascular
PDFBibTeX XMLCite
Full Text: DOI arXiv

References:

[1] Eck, V. G.; Donders, W. P.; Sturdy, J.; Feinberg, J.; Delhaas, T.; Hellevik, L. R.; Huberts, W., A guide to uncertainty quantification and sensitivity analysis for cardiovascular applications, Int. J. Numer. Methods Biomed. Eng., 32 (2015)
[2] Steinman, D. A.; Migliavacca, F., Editorial: Special issue on verification, validation, and uncertainty quantification of cardiovascular models: Towards effective VVUQ for translating cardiovascular modelling to clinical utilitys, Cardiovasc. Eng. Technol., 9 (2018)
[3] Sankaran, S.; Marsden, A. L., A stochastic collocation method for uncertainty quantification and propagation in cardiovascular simulations, J. Biomech. Eng., 133, 3, 1-12 (2011)
[4] Biehler, J.; Gee, M. W.; Wall, W. A., Towards efficient uncertainty quantification in complex and large-scale biomechanical problems based on a Bayesian multi-fidelity scheme, Biomech. Model. Mechanobiol., 14 (2014)
[5] Quicken, S.; Donders, W. P.; van Disseldorp, M. J.; Gashi, K.; Mees, B. M.E.; van de Vosse, F. N.; Lopata, R. G.P.; Delhaas, T.; Huberts, W., Application of an adaptive polynomial chaos expansion on computationally expensive three-dimensional cardiovascular models for uncertainty quantification and sensitivity analysis, J. Biomech. Eng., 138 (2016)
[6] Schiavazzi, D. E.; Doostan, A.; Iaccarino, G.; Marsden, A. L., A generalized multi-resolution expansion for uncertainty propagation with application to cardiovascular modeling, Comput. Methods Appl. Mech. Engrg., 314, 196-221 (2017) · Zbl 1439.74213
[7] Tran, J. S.; Schiavazzi, D. E.; Kahn, A. M.; Marsden, A. L., Uncertainty quantification of simulated biomechanical stimuli in coronary artery bypass grafts, Comput. Methods Appl. Mech. Engrg., 345, 402-428 (2019) · Zbl 1440.74233
[8] Seo, J.; Schiavazzi, D.; Kahn, A.; Marsden, A., The effects of clinically-derived parametric data uncertainty in patient-specific coronary simulations with deformable walls, Int. J. Numer. Methods Biomed. Eng., n/a, n/a, Article e3351 pp. (2020), arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1002/cnm.3351 URL https://onlinelibrary.wiley.com/doi/abs/10.1002/cnm.3351
[9] Colebank, M. J.; Paun, L. M.; Qureshi, M. U.; Chesler, M.; Husmeier, D.; Olufsen, M. S.; Fix, L. E., Influence of image segmentation on one-dimensional fluid dynamics predictions in the mouse pulmonary arteries, R. Soc. Interface (2019)
[10] Colebank, M. J.; Qureshi, M. U.; Olufsen, M. S., Sensitivity analysis and uncertainty quantification of 1-d models of pulmonary hemodynamics in mice under control and hypertensive conditions, Int. J. Numer. Methods Biomed. Eng. (2019)
[11] Boccadifuoco, A.; Mariotti, A.; Capellini, K.; Celi, S.; Salvetti, M. V., Uncertainty quantification applied to hemodynamic simulations of thoracic aorta aneurysms: Sensitivity to inlet conditions, Quantif. Uncertain.: Improv. Effic. Technol. (2020) · Zbl 1455.62209
[12] Fleeter, C.; Geraci, G.; Schiavazzi, D.; Kahn, A.; Marsden, A., Multilevel and multifidelity uncertainty quantification for cardiovascular hemodynamics, Comput. Methods Appl. Mech. Engrg., 365, Article 113030 pp. (2020) · Zbl 1442.76162
[13] Mirams, G. R.; Niederer, S. A.; Clayton, R. H., The fickle heart: Uncertainty quantification in cardiac and cardiovascular modeling and simulation, Philos. Trans. R. Soc. (2020)
[14] Campos, J. O.; Sundnes, J.; dos Santos, R. W.; Rocha, B. M., Uncertainty quantification and sensitivity analysis of left ventricular function during the full cardiac cycle, Philos. Trans. R. Soc. (2020)
[15] Clayton, R. H.; Aboelkassem, Y.; Cantwell, C. D.; Corrado, C.; Delhaas, T.; Huberts, W.; Lei, C. L.; Ni, H.; Panfilov, A. V.; Roney, C.; dos Santos, R. W., An audit of uncertainty in multi-scale cardiac electrophysiology models, Philos. Trans. R. Soc. (2020)
[16] Schiavazzi, D.; Arbia, G.; Baker, C.; Hlavacek, A.; Hsia, T.-Y.; Marsden, A.; Vignon-Clementel, I.; of Congenital Hearts Alliance (MOCHA) Investigators, T. M., Uncertainty quantification in virtual surgery hemodynamics predictions for single ventricle palliation, Int. J. Numer. Methods Biomed. Eng., 32, 3, Article e02737 pp. (2016)
[17] Tran, J.; Schiavazzi, D.; Ramachandra Bangalore, A.; Kahn, A.; Marsden, A., Automated tuning for parameter identification and uncertainty quantification in multi-scale coronary simulations, Comput. & Fluids, 142, 128-138 (2017) · Zbl 1390.76945
[18] Paun, L. M.; Colebank, M. J.; Olufsen, M. S.; Hill, N. A.; Husmeier, D., Assessing model mismatch and model selection in a bayesian uncertainty quantification analysis of a fluid-dynamics model of pulmonary blood circulation, J. R. Soc. Interface (2020)
[19] Sigal, I. A.; Hardisty, M. R.; Whyne, C. M., Mesh-morphing algorithms for specimen-specific finite element modeling, J. Biomech., 41 (2008)
[20] Le, M.; Unkelbach, J.; Ayache, N.; Delingette, H., Sampling image segmentations for uncertainty quantification, Med. Image Anal., 34 (2016)
[21] Joskowicz, L.; Cohen, D.; Caplan, N.; Sosna, J., Automatic segmentation variability estimation with segmentation priors, Med. Image Anal., 50 (2018)
[22] Gambaruto, A. M.; Janela, J.; Moura, A.; Sequeira, A., Sensitivity of hemodynamics in a patient specific cerebral aneurysm to vascular geometry and blood rehology, Math. Biosci. Eng. (2011) · Zbl 1259.76002
[23] Venugopal, P.; Li, X.; Cheng, L.; Bhagalia, R.; Edic, P. M., Sensitivity of FFR-CT to manual segmentation, Proc. SPIE (2019)
[24] Xu, H.; Baroli, D.; Veneziani, A., Global sensitivity analysis for patient-specific aortic simulations: the role of geometry, boundary condition and LES modeling parameters, Biomech. Eng. (2020)
[25] Sankaran, S.; Grady, L.; Taylor, C. A., Fast computation of hemodynamic sensitivity to lumen segmentation uncertainty, IEEE Trans. Med. Imaging, 34 (2015)
[26] Sankaran, S.; Grad, L. J.; Taylor, C. A., Real-time sensitivity analysis of blood flow simulations to lumen segmentation uncertainty, (Conference on Medical Imaging and Computer Assisted Intervention (2014))
[27] Sankaran, S.; Grady, L.; Taylor, C. A., Impact of geometric uncertainty on hemodynamic simulations using machine learning, Comput. Methods Appl. Mech. Eng., 297 (2015) · Zbl 1423.76530
[28] Sankaran, S.; Kim, H. J.; Choi, G.; Taylor, C. A., Uncertainty quantification in coronary blood flow simulations: impact of geometry, boundary conditions and blood viscosity, J. Biomech., 49 (2016)
[29] Gal, Y.; Ghahramani, Z., Dropout as a Bayesian approximation: representing model uncertainty in deep learning, (International Conference on Machine Learning (2016))
[30] Jena, R.; Awate, S. P., A Bayesian neural net to segment images with uncertainty estimates and good calibration, (International Conference on Information Processing in Imaging (2019))
[31] Nair, T.; Precup, D.; Arnold, D. L.; Arbel, T., Exploring uncertainty measures in deep networks for multiple sclerosis lesion detection and segmentation, Med. Image Anal., 59 (2020)
[32] Maher, G.; Parker, D.; Wilson, N.; Marsden, A., Neural network vessel lumen regression for automated segmentation in cardiovascular image-based modeling, Cardiovasc. Eng. Technol. (2020)
[33] Updegrove, A.; Wilson, N.; Merkow, J.; Lan, H.; Marsden, A.; Shadden, S., SimVascular: an open source pipeline for cardiovascular simulation, Ann. Biomed. Eng., 61 (2013)
[34] Wilson, N.; Ortiz, A.; Johnson, A., The vascular model repository: a public resource of medical imaging data and blood flow simulation results, J. Med. Devices, 7, 4 (2013)
[35] Updegrove, A.; Wilson, N. M.; Shadden, S. C., Boolean and smoothing of discrete surfaces, Adv. Eng. Softw., 95 (2016)
[36] Antiga, L.; Piccinelli, M.; Botti, L.; Ene-Iordache, B.; Remuzzi, A.; Steinman, D. A., An image-based modeling framework for patient-specific computational hemodynamics, Med. Biol. Eng. Comput., 46 (2008)
[37] Lesage, D.; Angelini, E. D.; Bloch, I.; Funka-Lea, G., A review of 3D vessel lumen segmentation techniques: Models, features and extraction schemes, Med. Image Anal., 13 (2009)
[38] Si, H., TetGen, a Delaunay-based quality tetrahedral mesh generator, ACM Trans. Math. Softw., 41 (2015) · Zbl 1369.65157
[39] LeCun, Y.; Bengio, Y.; Hinton, G., Deep learning, Nature, 521, 7553, 436-444 (2015)
[40] Schmidhuber, J., Deep learning in neural networks: an overview, Neural Netw., 61, 85-117 (2015)
[41] Hornik, K., Approximation capabilities of multilayer feedforward networks, Neural Netw., 4 (1991)
[42] Maas, A. L.; Hannun, A. Y.; Ng, A. Y., Rectifier nonlinearities improve neural network acoustic moels, ICML (2013)
[43] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, A. Rabinovich, Going deeper with convolutions, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 1-9.
[44] Srivastava, N.; Hinton, G.; Krizhevsky, A.; Sutskever, I.; Salakhutdinov, R., Dropout: A simple way to prevent neural networks from overfitting, J. Mach. Learn. Res. (2014) · Zbl 1318.68153
[45] He, K.; Zhang, X.; Ren, S.; Sun, J., Delving deep into rectifiers: surpassing human-level performance on ImageNet classification, (International Conference on Computer Vision (2015))
[46] Kingma, D.; Ba, J., Adam: A method for stochastic optimization, (International Conference on Learning Representations (2015))
[47] Brooks, A. N.; Hughes, T. J.R., Streamline upwind/Petrov-Galerkin formulations for convection dominated flows with particular emphasis on the incompressible Navier-Stokes equations, Comput. Methods Appl. Mech. Eng. (1982) · Zbl 0497.76041
[48] Moghadam, M. E.; Bazilevs, Y.; Hsia, T. Y.; Vignon-Clemente, I. E.; Marsden, A. L.; Alliance for the Modeling of Congenital Hearts Alliance, M., A comparsion of outlet boundary treatments for prevention of backflow divergence with relevance to blood flow simulations, Comput. Mech., 48 (2011) · Zbl 1398.76102
[49] Seo, J.; Schiavazzi, D.; Marsden, A., Performance of preconditioned iterative linear solvers for cardiovascular simulations in rigid and deformable vessels, Comput. Mech., 64, 3, 717-739 (2019) · Zbl 1465.74166
[50] Moghadam, M. E.; Vignon-Clementel, I. E.; Figliola, R.; Marsden, A. L.; Alliance for the Modeling of Congenital Hearts Alliance, M., A modular numerical method for implicit 0D/3D coupling in cardiovascular finite element simulations, J. Comput. Phys., 244 (2013) · Zbl 1377.76041
[51] Moghadam, M. E.; Bazilevs, Y.; Marsden, A. L., A bi-partitioned iterative algorithm for solving linear systems arising from incompressible flow problems, Comput. Methods Appl. Mech. Eng., 286 (2015) · Zbl 1423.76228
[52] Jansen, K. E.; Whiting, C. H.; Hulbert, G. M., A generalized-\( \alpha\) method for integrating the filtered Navier-Stokes equations with a stabilized finite element method, Comput. Methods Appl. Mech. Eng., 190 (2000) · Zbl 0973.76048
[53] Vignon-Clementel, I. E.; Figueroa, C. A.; Jansen, K. E.; Taylor, C. A., Outflow boundary conditions for 3D simulations of non-periodic blood flow and pressure fields in deformable arteries, Comput. Methods Biomech. Biomed. Eng., 13 (2009)
[54] Kim, H. J.; Vignon-Clementel, I. E.; Coogan, J. S.; Figueroa, C. A.; Jansen, K. E.; Taylor, C. A., Patient-specific modeling of blood flow and pressure in human coronary arteries, Ann. Biomed. Eng., 38 (2010)
[55] Baldi, P.; Sadowski, P., Understanding dropout, Adv. Neural Inf. Process. Syst., 26, 2814-2822 (2013)
[56] Guzzetti, S.; Alvarez, L. A.M.; Blanco, P. J.; Carlberg, K. T.; Veneziani, A., Propagating uncertainties in large-scale hemodynamics models via network uncertainty quantification and reduced-order modeling, Comput. Methods Appl. Mech. Eng., 358 (2020) · Zbl 1441.76147
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.