Version Franšaise

My work consists in the complete development of a non-contact diffuse optical tomography (DOT) scanner for small animal imaging. The goal is to be able to image the internal anatomy of small animals (mice) via the intrinsic optical properties of the biological tissues out of which they are composed. As well, the scanner will serve to image biomolecular processes taking place inside an animal via functionalized fluorescent biomarkers injected into it. In this case, we are talking about fluorescence DOT (FDOT). FDOT is a non-invasive biomedical imaging modality similarly to X-ray computerized tomography (CT), positron emission tomography (PET), or magnetic resonance imaging (MRI). FDOT, however, uses light and optical techniques, instead of ionizing radiation or radiofrequencies. My research covers all aspects needed for developing a scanner: hardware (optical, mechanical, and electronics design, along with phantom fabrication), software (automated scanner control, 3D computer vision for measuring the outer surface shape of the animal), and algorithms (light propagation modeling in the time domain, tomographic reconstruction algorithms for non-contact time-domain measurements, Monte Carlo simulations of light propagation). What distinguishes my work is the use of time-domain optical measurements using a laser emitting ultra-short pulses and ultra-fast time correlated single photon counting electronics. Moreover, measurements are made without contact with the animal. This will allow exploiting the information richness about the optical properties of a medium contained in time-domain signals. At the same time, this represents a great challenge, since then the tomographic reconstruction algorithms become more complex. Algorithm development is one of the aspects on which we are working intensely. Contrary to the other biomedical imaging modalities (CT, PET, MRI), there do not yet exist standard reconstruction algorithms in DOT and FDOT, the difficulty residing in the fact that light is strongly scattered in biological tissues, which makes the equations considerably more complicated. Molecular imaging using FDOT is a technology with a high potential for the pharmaceutical industry. Using fluorescent biomarkers, FDOT will eventually allow labelling new medications and drugs in order to follow their progression in whole animals to evaluate if they reach their targets. This will also allow following the treatment of a pathology on the same animal, and evaluate the efficacy of the treatment over long periods of time as required for longitudinal studies. This will allow eliminating variations from one animal to another, thereby getting more reliable statistics. Importantly, this will avoid the need sacrificing a large number of animals, as is required by current histological techniques. The screening of efficient new drugs will be accelerated, leading to the development of better medications faster and at lower cost. Another promising avenue for FDOT is oncology, where cancer cells can be identified in whole living subjects using fluorescent biomarkers that are capable of targeting specific membrane proteins expressed on the surface of such cells. This will lead to earlier detection of such cells, thereby improving patient outcomes.

My research on DOT and FDOT implies concepts related to physics (notably the understanding and the modelling of how light propagates in scattering media), medical imaging (tomographic reconstructions algorithms and instrumentation), and engineering (opto- electro- mechanical design and instrumentation development, electro-optics and electronics design).