This text is part of the special Quebec engineering booklet
Magnetic Resonance Imaging (MRI) is the tool neuroscience uses to examine and study the human brain. Can software engineering help to better interpret 3D images generated by MRI, through the writing of software and algorithms designed for this task?
This is the kind of problem that software engineer Sylvain Bouix tackled during his postdoctoral studies in Boston, at a hospital affiliated with Harvard University. Now a professor at the École de technologie supérieure (ETS), he has worked on developing a method for morphometric analysis of the brain.
“Morphometry is the study of shape, explains Sylvain Bouix, and in the case that concerns us, it is the study of the geometry and structures of the brain. We were looking for ways to better measure differences in brain morphometry. The work focused on the analysis of MRI images of two pathologies, namely schizophrenia and mild head trauma, such as cerebral concussion. “We were looking to detect structures in the brain that are abnormal and that could be linked to these two pathologies. My role was to develop computer tools capable of helping a neuroscientist better understand what is happening in the brain. »
A special route
At the start, nothing predestined Sylvain Bouix to take an interest in neuroscience. Passionate about science, he obtained his degree in software engineering in his native France with the aim of immediately entering the job market. “I first wanted to have an experience abroad, he continues, and I obtained an internship at the University of Kansas, where I met Professor John Gauch. It was the latter who introduced me to scientific research, which I had never thought of. He will complete a master’s degree in computer science at the University of Kansas, where he learns the basics of computer vision.
He then enrolled for a doctorate at McGill University and continued his work in computer vision, specializing in 3D shape analysis. “It was about developing algorithms representing three-dimensional shapes that a computer could understand,” he says. Professor Kaleem Siddiqi, who supervised his work, then allowed him to transpose his research into computer vision in the field of MRI images. The project involved measuring the change in the shape of the hippocampus in men and women over time. It was after his doctorate that we found him in Boston.
The choice of ETS
Several reasons explain why Sylvain Bouix chose to become a professor and researcher at the École de technologie supérieure. “I wanted to get closer to the engineering community,” he says. In Boston, I was immersed in the world of neuroscience, which is not my primary discipline. What better than an engineering school, like the ETS, to find myself surrounded by engineers whose concerns are similar to mine? »
And then, he missed teaching. “I’ve always loved teaching,” says Sylvain Bouix. And my conception of teaching is that of a two-way street. Obviously, teaching is the transfer of knowledge from teacher to student, but it is not only that. There are the students’ questions, the new perspectives that these questions open up, avenues that have sometimes been overlooked and which enrich the professor’s knowledge. It is this back and forth that fascinates me in graduate education. »
Sylvain Bouix has not for all that dismissed his interest in neuroscience, but the latter has developed and is no longer limited to the sole morphometric analysis of the human brain from MRI images. “Today, data in neuroscience comes to us from multiple sources,” he says. Of course, there are those from MRI image analyses, but there are also those from blood tests, cognitive tests and even questionnaires. This is an abundance of data that must be managed and formatted in order to make it more easily usable by neuroscientists. »
This special content was produced by the Special Publications team of the Duty, pertaining to marketing. The drafting of Duty did not take part.