Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
Retinal imaging is routinely used to provide information about vision health. Recent studies showed that retinal images can also provide information on general health (e.g. stress level), brain pathologies and attention. Depending on the target information, different imaging systems can be chosen and optimized. This thesis will focus on two distinct challenges in ophthalmology. The first one regards a see-through wearable ophthalmoscope for applications such as health monitoring and eye tracking. Such a system uses a scanning 2D micromirror that results in a compact confocal scanning laser ophthalmoscope. Another key feature of such a system is the use of a holographic plate on which a diffraction grating has been recorded. This allows the light from the laser beam to be diffracted to the eye without altering the patientâs field of view. Such a system is proposed and analyzed with simulations and with a proof of concept setup. The second challenge regards in-vivo imaging of retinal microstructures. Indeed, with the current state-of-the-art technology, it is still not possible to observe the morphology of transparent cellular structures in the top layers of the living retina with high contrast in a time frame relevant for use in patients. Such a possibility is appealing since neuronal layers of the retina could be used to obtain information on degenerative diseases of both retina and brain. In this thesis, it is shown how asymmetrical illumination of the retina results in phase contrast and how this can be used for fast imaging modalities of cellular structures. The technique is here detailed with a theoretical model, numerical simulations and ex-vivo experimental validations.
Christophe Moser, Timothé Laforest, Laura Emmanuelle Kowalczuk, Mathieu Künzi
Christophe Moser, Timothé Laforest, Laura Emmanuelle Kowalczuk, Dino Carpentras