The following thesis is collocated in the cardiovascular engineering field and the goal of this work is to compare different measures of pulse wave velocity and to study the effect of the elongation of the aorta during the contraction of the left ventricle. Arterial stiffness is a current topic of concern as it brings information about aortic function and conditions and it has been classified as an independent prognostic factor for cardiovascular and non cardiovascular diseases. Given that invasive techniques to measure aortic stiffness are risky and unfeasible on healthy subjects, some non-invasive measurements are commonly employed. The most robust and effective surrogate for the non-invasive evaluation of aortic stiffness is aortic pulse wave velocity. In this thesis the first focus was the comparison between different measurements compliance local, regional and global. In detail, it was investigated the coupling between 2D MRI FFT images and c-f PWV measures acquired from tonometry, to answer questions regarding the suitability of the c-f PWV as a measure of central compliance and its relation to the actual proximal aortic compliance obtained through 2D MRI, given that c-f PWV is an average metric and that the carotid-to-femoral path excludes the ascending aorta. The second focus of this work was to investigate the effect of the dynamic elongation of the proximal aorta during left ventricle contraction, evaluating its contribution to cross-sectional compliance. Furthermore, a correction model to account for the elongation of the ascending aorta when measuring the areal compliance was investigated. The evaluation of the different measures of pulse wave velocity is based on the processing of a large database of hemodynamic properties and data that have been acquired at the Geneva University Hospitals. The processing of data has been carried out on 3D Slicer, Vascular Modelling ToolKit (VMTK), MEDVISO software program Segment and RadiANT DICOM Viewer. The statistical analysis was performed using the data obtained from the post-processing in Matlab and Excel to the following aims: 1. Comparison of c-f PWV and of volumetric PWV with 2D MRI PWV; 2. Comparison of PWV calculated from cross-sectional area compliance (AAO) through Bramwell-Hill equation with reference method, before and after correcting it for elongation; 3. Development and evaluation of different regression models to predict PWV based on proximal aortic measurements and patient characteristics; 4. Understanding of how elongation is affected by various factors; 5. Comparison of PWV calculated from cross-sectional area compliance (DAO) through Bramwell-Hill equation with reference method. In agreement with previous literature, the main findings of this research are firstly that cross-sectional area changes are a better non-invasive estimator of the elasticity of the wall for the proximal ascending aorta with respect to the descending aorta. Then, pulse wave velocity of the ascending aorta turned out to be lower than the one of the proximal descending aorta, for all cases apart from pulse wave velocity calculated from volume compliance. Finally, it has been proved that there is a consistent age-dependent relation of pulse wave velocity, stronger for ascending aorta than for descending aorta. Despite these results, correlations between different measures of pulse wave velocity resulted to be not as strong as the ones of similar works. All in all, the sample of subjects involved in the study is still low to be able to run a correct evaluation and in a future prospective it has to be increased, eventually including diseased and elderly subjects as well.
La seguente tesi si colloca nel campo dell'ingegneria cardiovascolare e l'obiettivo di questo lavoro è quello di confrontare diverse misure di pulse wave velocity e di studiare l'effetto dell'elongazione dell'aorta durante la contrazione del ventricolo sinistro. La rigidezza arteriosa è un argomento di interesse attuale in quanto fornisce informazioni sulla funzione e sulle caratteristiche dell'aorta ed è stata classificata come fattore prognostico indipendente per le malattie cardiovascolari e non cardiovascolari. Dato che le tecniche invasive per misurare la rigidezza aortica sono rischiose e non praticabili su soggetti sani, comunemente sono usate alcune misurazioni non invasive. Il surrogato più robusto ed efficace per la valutazione non invasiva della rigidezza aortica è la pulse wave velocity. In questa tesi il primo obiettivo è stato il confronto tra diverse misure di compliance locale, regionale e globale. In particolare, è stato studiato il rapporto tra le immagini 2D MRI FFT e le misure c-f PWV acquisite tramite tonometria, per rispondere a domande riguardanti l'idoneità del c-f PWV come misura della compliance centrale e la sua relazione con l'effettiva compliance aortica prossimale ottenuta attraverso la risonanza magnetica 2D, dato che c-f PWV è una metrica media e che il percorso carotide-femorale esclude l'aorta ascendente. Il secondo obiettivo di questo lavoro è stato quello di indagare l'effetto dell'elongazione dinamica dell'aorta prossimale durante la contrazione del ventricolo sinistro, valutandone il contributo sulla compliance derivata da variazioni di area. Inoltre, è stato studiato un modello di correzione per tenere conto dell'elongazione dell'aorta ascendente durante la misurazione della compliance derivate da variazioni di area. La valutazione delle diverse misure di pulse wave velocity si basa sull'elaborazione di un ampio database di proprietà emodinamiche e di dati acquisiti presso gli Hôpitaux Universitaires de Genève. L'elaborazione dei dati è stata effettuata su 3D Slicer, Vascular Modelling ToolKit (VMTK), il software MEDVISO Segment e RadiANT DICOM Viewer. L'analisi statistica è stata effettuata utilizzando i dati ottenuti dalla post-elaborazione in Matlab e Excel con i seguenti scopi: 1. Confronto della c-f PWV e della PWV volumetrica con la 2D MRI PWV; 2. Confronto della PWV calcolata dalle variazioni di area trasversale (AAO) attraverso l'equazione di Bramwell-Hill con il metodo di riferimento, prima e dopo averla corretta per l’elongazione; 3. Sviluppo e valutazione di diversi modelli di regressione per prevedere la PWV sulla base di misurazioni aortiche prossimali e caratteristiche del paziente; 4. Studio di come l'elongazione sia influenzata da differenti fattori; 5. Confronto della PWV calcolata dalle variazioni di area trasversale (AAO) attraverso l'equazione di Bramwell-Hill con il metodo di riferimento. In accordo con la letteratura precedente, i principali risultati di questa ricerca sono innanzitutto che le variazioni di area della sezione trasversale sono un migliore indicatore non invasivo dell'elasticità della parete dell’'aorta ascendente rispetto all'aorta discendente prossimale. Secondariamente, la pulse wave velocity dell'aorta ascendente si è rivelata inferiore a quella dell'aorta discendente prossimale, per tutti i casi in studio ad eccezione della pulse wave velocity calcolata a partire da cambiamenti volumetrici. Infine, è stato dimostrato che esiste una relazione costante, dipendente dall'età, tra la pulse wave velocity e quella dell'aorta ascendente, più forte per l'aorta ascendente che per l'aorta discendente. Nonostante tali risultati, le correlazioni tra le diverse misure di pulse wave velocity non sono risultate tanto marcate quanto quelle di lavori simili. Nel complesso, il campione di soggetti coinvolti nello studio per eseguire una valutazione corretta è ancora ridotto e in una prospettiva futura deve essere aumentato, includendo eventualmente anche soggetti malati e anziani.
Comparison of different in vivo techniques to measure aortic wall compliance.
CALASTRA, CAMILLA GIULIA
2020/2021
Abstract
The following thesis is collocated in the cardiovascular engineering field and the goal of this work is to compare different measures of pulse wave velocity and to study the effect of the elongation of the aorta during the contraction of the left ventricle. Arterial stiffness is a current topic of concern as it brings information about aortic function and conditions and it has been classified as an independent prognostic factor for cardiovascular and non cardiovascular diseases. Given that invasive techniques to measure aortic stiffness are risky and unfeasible on healthy subjects, some non-invasive measurements are commonly employed. The most robust and effective surrogate for the non-invasive evaluation of aortic stiffness is aortic pulse wave velocity. In this thesis the first focus was the comparison between different measurements compliance local, regional and global. In detail, it was investigated the coupling between 2D MRI FFT images and c-f PWV measures acquired from tonometry, to answer questions regarding the suitability of the c-f PWV as a measure of central compliance and its relation to the actual proximal aortic compliance obtained through 2D MRI, given that c-f PWV is an average metric and that the carotid-to-femoral path excludes the ascending aorta. The second focus of this work was to investigate the effect of the dynamic elongation of the proximal aorta during left ventricle contraction, evaluating its contribution to cross-sectional compliance. Furthermore, a correction model to account for the elongation of the ascending aorta when measuring the areal compliance was investigated. The evaluation of the different measures of pulse wave velocity is based on the processing of a large database of hemodynamic properties and data that have been acquired at the Geneva University Hospitals. The processing of data has been carried out on 3D Slicer, Vascular Modelling ToolKit (VMTK), MEDVISO software program Segment and RadiANT DICOM Viewer. The statistical analysis was performed using the data obtained from the post-processing in Matlab and Excel to the following aims: 1. Comparison of c-f PWV and of volumetric PWV with 2D MRI PWV; 2. Comparison of PWV calculated from cross-sectional area compliance (AAO) through Bramwell-Hill equation with reference method, before and after correcting it for elongation; 3. Development and evaluation of different regression models to predict PWV based on proximal aortic measurements and patient characteristics; 4. Understanding of how elongation is affected by various factors; 5. Comparison of PWV calculated from cross-sectional area compliance (DAO) through Bramwell-Hill equation with reference method. In agreement with previous literature, the main findings of this research are firstly that cross-sectional area changes are a better non-invasive estimator of the elasticity of the wall for the proximal ascending aorta with respect to the descending aorta. Then, pulse wave velocity of the ascending aorta turned out to be lower than the one of the proximal descending aorta, for all cases apart from pulse wave velocity calculated from volume compliance. Finally, it has been proved that there is a consistent age-dependent relation of pulse wave velocity, stronger for ascending aorta than for descending aorta. Despite these results, correlations between different measures of pulse wave velocity resulted to be not as strong as the ones of similar works. All in all, the sample of subjects involved in the study is still low to be able to run a correct evaluation and in a future prospective it has to be increased, eventually including diseased and elderly subjects as well.| File | Dimensione | Formato | |
|---|---|---|---|
|
master_thesis16032021.pdf
solo utenti autorizzati dal 17/03/2022
Descrizione: Comparison of different in vivo techniques to measure aortic wall compliance
Dimensione
3.01 MB
Formato
Adobe PDF
|
3.01 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/10589/175322