Conception et caractérisation d'un outil de neuronavigation pour le rat

3D navigation is a concept that is used to find an object of interest using coordinates in space. Neuronavigation(NN) is the term used when the navigated space is within the confines of the nervous system, including brain. Neuronavigation systems are well-established in neurosurgery, especially in the human clinical field. There is also a growing presence in fundamental research which is widely based on animal studies. The rat is used as an important model in biomedical research. As of today, there is no commercial neuronavigation system on the market dedicated to small animal experiments. In neuroscientific studies of the rat, the localization of targets in the brain is taken from stereotaxic coordinates using paper atlases and specifically designed stereotaxic frames.

In this study, we developed a low-cost, high precision neuronavigation system dedicated to neuroscientific experiments in therat. More specifically, an ultrasound device was used to generate in vivo images of the rat skull which were then registered to a standard MRI image. Neuronavigation is based on the fusion between these two modalities; the ultrasound system conveying spatial information of the animal in the laboratory while being aligned to the precise anatomical image of the MRI.

In this thesis, both photoacoustic (PA) and ultrasound (US) imaging were used to generate in vivo data. In the first experiment, we constructed a home-made ultrasound transducer that was further characterized for photoacoustic use. Although it was able to generate good quality phantom images, no in vivo signal could be consistently recorded. A commercial transducer, looking at PA-US fusion imaging was then used in the next experiments. While PA did not provide useful information that could be used for in vivo neuronavigation, US provided good enough image quality to perform registration with the MRI.

The accuracy of the neuronavigation system reported here was characterized in a post mortemstudy. The accuracy of the neuronavigation device was defined by the distance between a selected target in the standard MRI brain and an injection site calculated from the injection zone measured on the histological slices. Results showed that the system had a mean error of 1.13mm, with a minimal error value of 500μm. These numbers attest to the potential of this neuronavigation method.