Abstract
Objective
In the management of the aortic aneurysm, 4D flow magnetic resonance Imaging provides valuable information for the computation of new biomarkers using computational fluid dynamics (CFD). However, accurate segmentation of the aorta is required. Thus, our objective is to evaluate the performance of two automatic segmentation methods on the calculation of aortic wall pressure.
Methods
Automatic segmentation of the aorta was performed with methods based on deep learning and multi-atlas using the systolic phase in the 4D flow MRI magnitude image of 36 patients. Using mesh morphing, isotopological meshes were generated, and CFD was performed to calculate the aortic wall pressure. Node-to-node comparisons of the pressure results were made to identify the most robust automatic method respect to the pressures obtained with a manually segmented model.
Results
Deep learning approach presented the best segmentation performance with a mean Dice similarity coefficient and a mean Hausdorff distance (HD) equal to 0.92+/− 0.02 and 21.02+/− 24.20 mm, respectively. At the global level HD is affected by the performance in the abdominal aorta. Locally, this distance decreases to 9.41+/− 3.45 and 5.82+/− 6.23 for the ascending and descending thoracic aorta, respectively. Moreover, with respect to the pressures from the manual segmentations, the differences in the pressures computed from deep learning were lower than those computed from multi-atlas method.
Conclusion
To reduce biases in the calculation of aortic wall pressure, accurate segmentation is needed, particularly in regions with high blood flow velocities. Thus, the deep learning segmen-tation method should be preferred.
Similar content being viewed by others
References
Pinard A, Jones GT, Milewicz DM (2019) Genetics of thoracic and abdominal aortic diseases: aneurysms, dissections, and ruptures. Circ Res 124(4):588–606
Kuzmik GA, Sang AX, Elefteriades JA (2012) Natural history of thoracic aortic aneurysms. J Vasc Surg 56(2):565–571
Ehrman JK, Fernandez AB, Myers J, Oh P, Thompson PD, Keteyian SJ (2020) Aortic aneurysm: diagnosis, management, exercise testing, and training. J Cardiopulm Rehabil Prev 40(4):215–223
Adamo L, Braverman AC (2015) Surgical threshold for bicuspid aortic valve aneurysm: a case for individual decision-making. Heart 101(17):1361–1367
Pape L, Tsai T, Isselbacher E, Oh J, O’gara P, Evangelista A, Fattori R, Meinhardt G, Trimarchi S, Bossone E et al (2007) International registry of acute aortic dissection (IRAD) investigators: aortic diameter \(>\) or = 5.5 cm is not a good predictor of type a aortic dissection: observations from the international registry of acute aortic dissection (IRAD). Circulation 116:1120–1127
Condemi F, Campisi S, Viallon M, Croisille P, Avril S (2019) Relationship between ascending thoracic aortic aneurysms hemodynamics and biomechanical properties. IEEE Trans Biomed Eng 67(4):949–956
Stankovic Z, Allen BD, Garcia J, Jarvis KB, Markl M (2014) 4d flow imaging with MRI. Cardiovasc Diagn Therapy 4(2):173
Ha H, Kim GB, Kweon J, Lee SJ, Kim Y-H, Lee DH, Yang DH, Kim N (2016) Hemodynamic measurement using four-dimensional phase-contrast MRI: quantification of hemodynamic parameters and clinical applications. Korean J Radiol 17(4):445–462
van Pelt R, Nguyen H, ter Haar Romeny B, Vilanova A (2012) Automated segmentation of blood-flow regions in large thoracic arteries using 3D-cine PC-MRI measurements. Int J Comput Assist Radiol Surg 7(2):217–224
Berhane H, Scott M, Elbaz M, Jarvis K, McCarthy P, Carr J, Malaisrie C, Avery R, Barker AJ, Robinson JD et al (2020) Fully automated 3D aortic segmentation of 4D flow MRI for hemodynamic analysis using deep learning. Magn Reson Med 84(4):2204–2218
Fujiwara T, Berhane H, Scott MB, Englund EK, Schäfer M, Fonseca B, Berthusen A, Robinson JD, Rigsby CK, Browne LP et al (2022) Segmentation of the aorta and pulmonary arteries based on 4D flow MRI in the pediatric setting using fully automated multi-site, multi-vendor, and multi-label dense u-net. J Magn Reson Imaging 55(6):1666–1680
Bustamante M, Gupta V, Forsberg D, Carlhäll C-J, Engvall J, Ebbers T (2018) Automated multi-atlas segmentation of cardiac 4D flow MRI. Med Image Anal 49:128–140
Bustamante M, Viola F, Engvall J, Carlhäll C-J, Ebbers T (2023) Automatic time-resolved cardiovascular segmentation of 4D flow MRI using deep learning. J Magn Reson Imaging 57(1):191–203
Leuprecht A, Kozerke S, Boesiger P, Perktold K (2003) Blood flow in the human ascending aorta: a combined MRI and CFD study. J Eng Math 47(3):387–404
Numata S, Itatani K, Kanda K, Doi K, Yamazaki S, Morimoto K, Manabe K, Ikemoto YHK (2016) Blood flow analysis of the aortic arch using computational fluid dynamics. Eur J Cardio-Thorac Surg 49(6):1578–1585
Gülan U, Calen C, Duru F, Holzner M (2018) Blood flow patterns and pressure loss in the ascending aorta: a comparative study on physiological and aneurysmal conditions. J Biomech 76:152–159
Renner J, Nadali Najafabadi H, Modin D, Länne T, Karlsson M (2012) Subject-specific aortic wall shear stress estimations using semi-automatic segmentation. Clin Physiol Funct Imaging 32(6):481–491
Alexa M (2002) Recent advances in mesh morphing. Comput Graph Forum 21(2):173–198
Geronzi L, Gasparotti E, Capellini K, Cella U, Groth C, Porziani S, Chiappa A, Celi S, Biancolini ME (2021) High fidelity fluid-structure interaction by radial basis functions mesh adaption of moving walls: a workflow applied to an aortic valve. J Comput Sci 51:101327
Biancolini ME (2012) Mesh morphing and smoothing by means of radial basis functions (RBF): a practical example using fluent and RBF morph. In: Handbook of research on computational science and engineering: theory and practice. IGI Global, pp 347–380
Maquart T, Wenfeng Y, Elguedj T, Gravouil A, Rochette M (2020) 3D volumetric isotopological meshing for finite element and isogeometric based reduced order modeling. Comput Methods Appl Mech Eng 362:112809
Biancolini ME, Ponzini R, Antiga L, Morbiducci U (2012) A new workflow for patient specific image-based hemodynamics: parametric study of the carotid bifurcation. Model objects represented images III. Fundam Methods Appl Comput
Capellini K, Vignali E, Costa E, Gasparotti E, Biancolini ME, Landini L, Positano V, Celi S (2018) Computational fluid dynamic study for ATAA hemodynamics: an integrated image-based and radial basis functions mesh morphing approach. J Biomech Eng 140(11):111007
Capellini K, Gasparotti E, Cella U, Costa E, Fanni BM, Groth C, Porziani S, ME B, Celi S, (2021) A novel formulation for the study of the ascending aortic fluid dynamics with in vivo data. Med Eng Phys 91:68–78
Groth C, Porziani S, Biancolini ME, Costa E, Celi S, Capellini K, Rochette M, Morgenthaler V (2018) The medical digital twin assisted by reduced order models and mesh morphing 10
Indrakusuma R, Jalalzadeh H, Planken R, Marquering H, Legemate D, Koelemay M, Balm R (2016) Biomechanical imaging markers as predictors of abdominal aortic aneurysm growth or rupture: a systematic review. Eur J Vasc Endovasc Surg 52(4):475–486
Catalano C, Agnese V, Gentile G, Raffa GM, Pilato M, Pasta S (2021) Atlas-based evaluation of hemodynamic in ascending thoracic aortic aneurysms. Appl Sci 12(1):394
Pasta S, Gentile G, Raffa G, Bellavia D, Chiarello G, Liotta R, Luca A, Scardulla C, Pilato M (2017) In silico shear and intramural stresses are linked to aortic valve morphology in dilated ascending aorta. Eur J Vasc Endovasc Surg 54(2):254–263
Iglesias JE, Sabuncu MR (2015) Multi-atlas segmentation of biomedical images: a survey. Med Image Anal 24(1):205–219
Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13(4):600–612
Marin-Castrillon DM, Boucher A, Lin S, Bernard C, Morgant M-C, Cochet A, Lalande A, Bouchot O, Presles B (2021) Multi-atlas segmentation of the aorta from 4D flow MRI: comparison of several fusion strategies. In: International workshop on statistical atlases and computational models of the heart. Springer, pp 3–11
Artaechevarria X, Munoz-Barrutia A, Ortiz-de-Solorzano C (2009) Combination strategies in multi-atlas image segmentation: application to brain MR data. IEEE Trans Med Imaging 28(8):1266–1277
Ronneberger O, Fischer P, Brox T (2015) U-net: Convolutional networks for biomedical image segmentation. In: International conference on medical image computing and computer-assisted intervention. Springer, pp 234–241
Çiçek Ö, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O (2016) 3D U-net: learning dense volumetric segmentation from sparse annotation. In: International conference on medical image computing and computer-assisted intervention. Springer, pp 424–432
Badrinarayanan V, Kendall A, Cipolla R (2017) Segnet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans Pattern Anal Mach Intell 39(12):2481–2495
Janssens R, Zeng G, Zheng G (2018) Fully automatic segmentation of lumbar vertebrae from CT images using cascaded 3D fully convolutional networks. In: 2018 IEEE 15th international symposium on biomedical imaging (ISBI 2018). IEEE, pp 893–897
Grassi L, Hraiech N, Schileo E, Ansaloni M, Rochette M, Viceconti M (2011) Evaluation of the generality and accuracy of a new mesh morphing procedure for the human femur. Med Eng Phys 33(1):112–120
Buhmann MD (2000) Radial basis functions. Acta Numer 9:1–38
Braaten M, Shyy W (1986) A study of recirculating flow computation using body-fitted coordinates: consistency aspects and mesh skewness. Numer Heat Transf Part A Appl 9(5):559–574
Shore A (2000) Capillaroscopy and the measurement of capillary pressure. Br J Clin Pharmacol 50(6):501–513
Liang L, Mao W, Sun W (2020) A feasibility study of deep learning for predicting hemodynamics of human thoracic aorta. J Biomech 99:109544
Warfield SK, Zou KH, Wells WM (2004) Simultaneous truth and performance level estimation (staple): an algorithm for the validation of image segmentation. IEEE Trans Med Imaging 23(7):903–921
Lalande A, Garreau M, Frouin F (2015) Multi-modality cardiac imaging: processing and analysis. Wiley, New York, pp 184–192 (Chap. 5)
Klein S, Van Der Heide UA, Lips IM, Van Vulpen M, Staring M, Pluim JP (2008) Automatic segmentation of the prostate in 3D MR images by atlas matching using localized mutual information. Med Phys 35(4):1407–1417
Krygier MC, LaBonte T, Martinez C, Norris C, Sharma K, Collins LN, Roberts SA (2021) Quantifying the unknown impact of segmentation uncertainty on image-based simulations. Nat Commun 12(1):1–11
Brüning J, Hellmeier F, Yevtushenko P, Kühne T, Goubergrits L (2018) Uncertainty quantification for non-invasive assessment of pressure drop across a coarctation of the aorta using CFD. Cardiovasc Eng Technol 9(4):582–596
Maher GD, Fleeter CM, Schiavazzi DE, Marsden AL (2021) Geometric uncertainty in patient-specific cardiovascular modeling with convolutional dropout networks. Comput Methods Appl Mech Eng 386
Bryan R, Mohan PS, Hopkins A, Galloway F, Taylor M, Nair PB (2010) Statistical modelling of the whole human femur incorporating geometric and material properties. Med Eng Phys 32(1):57–65
Porziani S, Groth C, Waldman W, Biancolini ME (2021) Automatic shape optimisation of structural parts driven by BGM and RBF mesh morphing. Int J Mech Sci 189:105976
Skala V (2017) RBF interpolation with CSRBF of large data sets. Procedia Comput Sci 108:2433–2437
Hoeijmakers M, Huberts W, Rutten M, van de Vosse F (2021) The impact of shape uncertainty on aortic-valve pressure-drop computations. Int J Numer Methods Biomed Eng 37(10):3518
Karmonik C, Diaz O, Klucznik R, Grossman RG, Zhang YJ, Britz G, Lv N, Huang Q (2015) Quantitative comparison of hemodynamic parameters from steady and transient CFD simulations in cerebral aneurysms with focus on the aneurysm ostium. J Neurointerv Surg 7(5):367–372
Condemi F, Campisi S, Viallon M, Troalen T, Xuexin G, Barker A, Markl M, Croisille P, Trabelsi O, Cavinato C et al (2017) Fluid-and biomechanical analysis of ascending thoracic aorta aneurysm with concomitant aortic insufficiency. Ann Biomed Eng 45(12):2921–2932
Gallo D, Gülan U, Di Stefano A, Ponzini R, Lüthi B, Holzner M, Morbiducci U (2014) Analysis of thoracic aorta hemodynamics using 3D particle tracking velocimetry and computational fluid dynamics. J Biomech 47(12):3149–3155
Reymond P, Crosetto P, Deparis S, Quarteroni A, Stergiopulos N (2013) Physiological simulation of blood flow in the aorta: comparison of hemodynamic indices as predicted by 3-D FSI, 3-D rigid wall and 1-D models. Med Eng Phys 35(6):784–791
Pons R, Guala A, Rodríguez-Palomares JF, Cajas J, Dux-Santoy L, Teixidó-Tura G, Molins JJ, Vázquez M, Evangelista A, Martorell J (2020) Fluid-structure interaction simulations outperform computational fluid dynamics in the description of thoracic aorta haemodynamics and in the differentiation of progressive dilation in Marfan syndrome patients. Roy Soc Open Sci 7(2):191752
Suito H, Takizawa K, Huynh VQ, Sze D, Ueda T (2014) FSI analysis of the blood flow and geometrical characteristics in the thoracic aorta. Comput Mech 54(4):1035–1045
Mendez V, Di Giuseppe M, Pasta S (2018) Comparison of hemodynamic and structural indices of ascending thoracic aortic aneurysm as predicted by 2-way FSI, CFD rigid wall simulation and patient-specific displacement-based FEA. Comput Biol Med 100:221–229
Acknowledgements
We would like to thank Siemens Healthineers Company for providing the 4D flow MRI sequence.
Funding
The authors did not receive support from any organization for the submitted work.
Author information
Authors and Affiliations
Contributions
DMM-C: Methodology, Software, Conceptualization, Data Curation, Writing—Original Draft, Visualization, Investigation, Formal analysis. LG: Methodology, Software, Conceptualization, Writing—Original Draft, Visualization. AB: Supervision, Data Curation, Writing—Review and Editing. SL: Resources, Investigation. M-CM: Resources Alexandre Cochet: Writing—Review and Editing, Methodology. MR: Funding acquisition, Writing—Review and Editing, Methodology. SL: Conceptualization, Writing—Review and Editing. KA: Review and Editing. NJ: Review and Editing. LSA: Data Curation, Formal analysis. AL: Conceptualization, Writing—Review and Editing, Resources. OB: Supervision, Project administration. BP: Supervision, Conceptualization, Writing—Review and Editing, Validation.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare they have no conflict of interest. Financial interest: The authors have no relevant financial or non-financial interests to disclose.
Ethical standards
We declare that this research did not received any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Marin-Castrillon, D.M., Geronzi, L., Boucher, A. et al. Segmentation of the aorta in systolic phase from 4D flow MRI: multi-atlas vs. deep learning. Magn Reson Mater Phy 36, 687–700 (2023). https://doi.org/10.1007/s10334-023-01066-2
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10334-023-01066-2