DescriptionThe iMed project is about the design of a method to automatically detect emboli in the vessel tree of the pulmonary artery, from HCT (helicoidal computed tomography) millimeter slices.
To achieve this, the radiologist must realize a specific exam (angiography) to opacify the artery. A contrast product is used with appropriate quantity and timing so that the artery appears clearly (not the veins). On these images, healthy sections appear opaque (white) while emboli appear as fuzzy, darker spots.
The algorithms should first allow to segment the vessel tree and next proceed to analyse their content through well chosen criteria. Another challenge is the huge amount of data peculiar to high resolution volume (3D) imaging.
This subject is unique in the sense it is new and scientifically challenging through health and social objectives.
PeopleRaphaël Sebbe, TCTS, FPMs, is pursuing a PhD thesis for this project, under the responsibility of Bernard Gosselin