Fluid-structure interaction (FSI) is a sophisticated analysis used in the cardiovascular field to investigate the hemodynamics and the interaction between blood flow and the vessels walls. This analysis is typically solved by a 2-way approach, extremely accurate but with a high computational cost. On the other hand, Coupled Momentum Method for Fluid-Solid Interaction (CMM-FSI) is an innovative approach for solving FSI that involves the modeling of the wall as a flexible membrane. In this way, it is possible to take into account the vessel deformation, greatly simplifying the solution of the analytical problem. The main purpose of this thesis is to compare the results obtained by exploiting the two different solving methods, implemented in three different commercial software, CRIMSON, SimVascular, and ANSYS. We chose as a case study a patient who underwent valve-sparing reconstruction and investigated biomechanical changes from pre- to post-operation conditions, induced by graft implantation. Such alterations, caused by the mechanical properties and geometry of the synthetic graft can lead to the pathogenesis of diseases such as atherosclerosis and dissection. The anatomy of the patient was reconstructed from angiographic MRI images. Velocities and flow rates for inlet and outlet conditions were derived from pre-and postoperative 4D-Flow data, allowing the development of a patient-tailored model. Results were validated against 4D flow velocity data, obtaining a mean RRMSE value below 0.235±0.067. For the comparison between solvers, commonly investigated hemodynamic indexes (e.g., WSS, TAWSS, OSI) and wall strain, were evaluated and compared i) in the same surgical configuration between different solvers ii) in different surgical configurations (pre- vs. post-) for the same solver. For the latter, all solutions identified a significant increase in the indexes of interest (p<0.05) in the descending aorta except for the strain, where the CMM-FSI simulations detected a significant decrease (p<0.001) while the 2-way point out an opposite outcome. In conclusion, all methods tested were able to assess hemodynamics but to obtain accurate structural results use of 2-way FSI is preferable.
L'interazione fluido-struttura (FSI) è un'analisi usata nel campo cardiovascolare per studiare l'emodinamica e l'interazione tra il flusso sanguigno e le pareti dei vasi. Questa analisi è tipicamente risolta con un approccio 2-way, estremamente accurato ma dall'alto costo computazionale. Viceversa, il Coupled Momentum Method for Fluid-Solid Interaction (CMM-FSI) è modo innovativo per risolvere le FSI che prevede la modellazione della parete come membrana flessibile. In questo modo, è possibile considerare la deformazione del vaso, semplificando notevolmente la soluzione del problema analitico. Lo scopo principale di questa tesi è quello di confrontare i risultati ottenuti con i due diversi metodi risolutivi, implementati in tre software commerciali, CRIMSON, SimVascular e ANSYS. Come caso studio si è scelto un paziente sottoposto a ricostruzione valve sparing e sono stati indagati i cambiamenti biomeccanici dalle condizioni pre- a post-operatorie. Tali alterazioni, causate dalle proprietà meccaniche e dalla geometria del graft sintetico, possono portare alla patogenesi di malattie come l'aterosclerosi e la dissezione. L'anatomia del paziente è stata ricostruita da immagini angiografiche MRI. Le velocità e le portate per le condizioni di ingresso e uscita sono state derivate dai dati 4D-Flow, consentendo lo sviluppo di un modello paziente-specifico. I risultati sono stati validati rispetto ai dati di velocità 4D-Flow, ottenendo un valore medio di RRMSE inferiore a 0.235±0.067. Per il confronto tra i solutori, gli indici emodinamici comunemente studiati (e.g., Wall Shear Stress, Time-Averaged WSS, Oscillatory Shear Index) e lo strain in parete, sono stati valutati e confrontati i) nella stessa configurazione chirurgica tra diversi solutori ii) in diverse configurazioni chirurgiche (pre vs post) per lo stesso solutore. Per quest'ultima analisi, tutte le soluzioni hanno identificato un aumento significativo degli indici di interesse (p<0,05) in aorta discendente, tranne che per lo strain, dove le simulazioni CMM-FSI hanno rilevato una diminuzione significativa (p<0,001), mentre le 2-way un risultato opposto. In conclusione, tutti i metodi testati sono stati in grado di valutare l'emodinamica vascolare ma per ottenere risultati strutturali accurati è preferibile l'uso di FSI 2-way.
Assessment of ascending thoracic aortic aneurysm prior and after valve-sparing surgery : tailored analysis comparing the Coupled Momentum Method and the 2-way Fluid-Solid Interaction
Saffioti, Alice
2020/2021
Abstract
Fluid-structure interaction (FSI) is a sophisticated analysis used in the cardiovascular field to investigate the hemodynamics and the interaction between blood flow and the vessels walls. This analysis is typically solved by a 2-way approach, extremely accurate but with a high computational cost. On the other hand, Coupled Momentum Method for Fluid-Solid Interaction (CMM-FSI) is an innovative approach for solving FSI that involves the modeling of the wall as a flexible membrane. In this way, it is possible to take into account the vessel deformation, greatly simplifying the solution of the analytical problem. The main purpose of this thesis is to compare the results obtained by exploiting the two different solving methods, implemented in three different commercial software, CRIMSON, SimVascular, and ANSYS. We chose as a case study a patient who underwent valve-sparing reconstruction and investigated biomechanical changes from pre- to post-operation conditions, induced by graft implantation. Such alterations, caused by the mechanical properties and geometry of the synthetic graft can lead to the pathogenesis of diseases such as atherosclerosis and dissection. The anatomy of the patient was reconstructed from angiographic MRI images. Velocities and flow rates for inlet and outlet conditions were derived from pre-and postoperative 4D-Flow data, allowing the development of a patient-tailored model. Results were validated against 4D flow velocity data, obtaining a mean RRMSE value below 0.235±0.067. For the comparison between solvers, commonly investigated hemodynamic indexes (e.g., WSS, TAWSS, OSI) and wall strain, were evaluated and compared i) in the same surgical configuration between different solvers ii) in different surgical configurations (pre- vs. post-) for the same solver. For the latter, all solutions identified a significant increase in the indexes of interest (p<0.05) in the descending aorta except for the strain, where the CMM-FSI simulations detected a significant decrease (p<0.001) while the 2-way point out an opposite outcome. In conclusion, all methods tested were able to assess hemodynamics but to obtain accurate structural results use of 2-way FSI is preferable.| File | Dimensione | Formato | |
|---|---|---|---|
|
Saffioti_Tesi_finale.pdf
solo utenti autorizzati dal 13/04/2023
Descrizione: Tesi
Dimensione
23.22 MB
Formato
Adobe PDF
|
23.22 MB | Adobe PDF | Visualizza/Apri |
|
Saffioti_Executive_Summary.pdf
solo utenti autorizzati dal 13/04/2023
Descrizione: Sommario esteso
Dimensione
2 MB
Formato
Adobe PDF
|
2 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/188468