| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.44 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
A retina representa o único tecido no qual os vasos sanguíneos podem ser visualizados in
vivo, de forma não invasiva. Desta forma, o estudo das suas alterações, causadas pela presença
de doenças oculares, permite averiguar outros processos patológicos que estejam presentes.
Médicos e investigadores pressupõem haver uma relação entre a ocorrência de acidentes
vasculares, incluindo o acidente vascular cerebral, e alterações na vasculatura e hemodinâmica da
retina ao nível do seu plexo vascular profundo. Técnicas de imagem não invasivas permitem a
visualização da vasculatura ocular, mas não conseguem ser totalmente informativas do
comportamento do fluxo sanguíneo, ao longo do desenvolvimento de patologias. Assim, a junção
da imagiologia com um estudo de simulação complementar, é necessário para uma análise mais
detalhada.
Apesar dos estudos de fluidodinâmica computacional do fluxo sanguíneo na retina
realizados não serem abundantes, a maioria tem sido realizada em contexto bidimensional, não
sendo considerada a tridimensionalidade dos vasos ou a espacialidade ao longo dos diferentes
plexos vasculares. Para além disso, nenhum estudo de análise computacional foi feito no plexo
profundo da retina.
Este trabalho tem como principal objetivo a criação de uma metodologia que permita a
análise computacional do fluxo sanguíneo na microvasculatura da retina, a partir de imagens 3D.
As etapas principais desta metodologia consistem no processamento das imagens 3D, na criação
da malha de superfície e da malha interna da estrutura e, por fim, na simulação de fluxo de volume
finito.
Os resultados obtidos de distribuição de velocidade e pressão, mostraram-se de acordo
com a literatura. A metodologia desenvolvida apresentou ser computacionalmente exequível, com
resultados bastante robustos ao nível da geometria estrutural. No entanto, são necessários estudos
mais aprofundados relativamente à criação da malha de superfície, de forma a evitar retificações
manuais complexas, e poder-se alargar a metodologia deste trabalho a toda a rede vascular
retiniana.
The retina represents the only tissue in which blood vessels can be visualized in vivo, non-invasively. In this way, the study of its alterations, caused by the presence of ocular diseases, allows for the investigation of other pathological processes that may be present. Physicians and researchers assume that there is a relationship between the occurrence of vascular accidents, including stroke, and changes in the vasculature and hemodynamics of the retina at the level of its deep vascular plexus. Non-invasive imaging techniques allow for the visualization of ocular vasculature, but they cannot provide a complete understanding of blood flow behavior throughout the development of pathologies. Thus, combining imaging with complementary simulation studies is necessary for a more detailed analysis. Despite the limited number of computational fluid dynamics studies of blood flow in the retina, most of them have been conducted in a two-dimensional context, without considering the three-dimensionality of vessels or spatiality along different vascular plexuses. Additionally, no computational analysis study has been conducted on the deep retinal plexus. The main objective of this work is to create a methodology that allows for the computational analysis of blood flow in the retinal microvasculature from 3D images. The main steps of this methodology involve processing the 3D images, creating the surface mesh and internal mesh of the structure, and finally, simulating finite volume flow. The results obtained for velocity and pressure distribution were consistent with the literature. The developed methodology proved to be computationally feasible, with robust results at the structural geometry level. However, further studies are needed regarding the creation of the surface mesh to avoid complex manual adjustments and to extend the methodology of this work to the entire retinal vascular network.
The retina represents the only tissue in which blood vessels can be visualized in vivo, non-invasively. In this way, the study of its alterations, caused by the presence of ocular diseases, allows for the investigation of other pathological processes that may be present. Physicians and researchers assume that there is a relationship between the occurrence of vascular accidents, including stroke, and changes in the vasculature and hemodynamics of the retina at the level of its deep vascular plexus. Non-invasive imaging techniques allow for the visualization of ocular vasculature, but they cannot provide a complete understanding of blood flow behavior throughout the development of pathologies. Thus, combining imaging with complementary simulation studies is necessary for a more detailed analysis. Despite the limited number of computational fluid dynamics studies of blood flow in the retina, most of them have been conducted in a two-dimensional context, without considering the three-dimensionality of vessels or spatiality along different vascular plexuses. Additionally, no computational analysis study has been conducted on the deep retinal plexus. The main objective of this work is to create a methodology that allows for the computational analysis of blood flow in the retinal microvasculature from 3D images. The main steps of this methodology involve processing the 3D images, creating the surface mesh and internal mesh of the structure, and finally, simulating finite volume flow. The results obtained for velocity and pressure distribution were consistent with the literature. The developed methodology proved to be computationally feasible, with robust results at the structural geometry level. However, further studies are needed regarding the creation of the surface mesh to avoid complex manual adjustments and to extend the methodology of this work to the entire retinal vascular network.
Descrição
Tese de mestrado, Engenharia Biomédica e Biofísica , 2023, Universidade de Lisboa, Faculdade de Ciências
Palavras-chave
Modelo 3D Retina Análise CFD Fluxo Sanguíneo Plexo vascular profundo Teses de mestrado - 2024
