Segmentation et extraction de caractéristiques des vaisseaux sanguins cérébraux à l'aide de l'IRM

The neurovascular coupling is a growing field; it studies the effects of cerebral activity on the behaviour of cerebral blood flow (CBF) and the blood vessels themselves. With the help of magnetic resonance imaging (MRI), it is possible to obtain images such as susceptibility weighted imaging (SWI) to see the veins or time-of-flight magnetic resonance angiography (TOF MRA) to visualize the arteries. These images allow having a structural representation of vessels in the brain. This thesis presents a method to segment blood vessels from structural images and extract their features. Using the segmentation mask, it is possible to calculate the diameter of the vessels as well as their length. With the help of such automatic segmentation tools, we conducted a study to analyze the behaviour of blood vessels during neuronal activities. Due to visual stimulation, we have acquired two images; one at rest and the other with stimulation. We compare the diameter in each of the images and obtain vasodilation in millimeters, but also as a percentage in each voxel. We also calculated the distance between the activation site and each voxel to see the magnitude of the vasodilation function of the distance. All this provides a better understanding of the vascular system of the human brain.

Name thesis:degree_name

M. Sc.

Level thesis:degree_level

Maîtrise

Discipline thesis:degree_discipline

Sciences des radiations et imagerie biomédicale

Grantor dc:publisher

Université de Sherbrooke

Year dc:date.issued

2017

Author dc:creator

• Bizeau, Alexandre

Advisors dc:contributor.advisor

• Whittingstall, Kevin
• Descoteaux, Maxime
• Bocti, Christian

• Temps de vol (Time-of-flight)
• Diamètre
• Stimulation
• Volume sanguin cérébral
• Flux sanguin cérébral
• Time-of-flight
• Diameter
• Cerebral blood volume
• Cerebral blood flow

Licence dc:rights.uri

• http://creativecommons.org/licenses/by-nc-nd/2.5/ca/

Language dc:language.iso

fr, en

Handle dc:identifier.uri

http://hdl.handle.net/11143/10259

OAI identifier oai:identifier

oai:usherbrooke.scholaris.ca:11143/10259