Hello to the whole SciGi family, my name is Elina THIBEAU and I am in post-doctorate since February 15, 2022 and for two years at the University of Twente in the Netherlands.
My thesis helped constitute and develop methods for the study of Alzheimer’s disease, the leading cause of dementia in the world.
There are indeed still many studies of this disease, because despite its definition more than a century ago, disagreement persists about the triggering mechanisms and ways to slow its progression.
A thesis useful to society
Today, my colleagues and I are working to develop methods to help physicians better understand and diagnose the various forms of this disease. These methods called deep learning are actually a type of artificial intelligence made up of networks of artificial neurons inspired by the human brain. The system can then learn on its own in time. As these new methods are being developed, there are still problems. In the field of Alzheimer’s study, there is a real problem of reproducibility of experiments carried out with these methods: thus two research teams carrying out the same experiment can obtain very different results, which means we can’t come to any conclusions.
Thanks to my thesis entitled: Reproducible and interpretable deep learning for the diagnosis, prognosis, and subtyping of Alzheimer’s disease from neuroimaging, I was able to become an official researcher. I showed the flaws in the methods used that led to this lack of reproducibility, helped to better interpret the results of these methods and produced a code base that allows researchers to conduct experiments in an environment that prevents these flaws.
I started in my laboratory with a research internship of 6 months. Even though it is not part of the thesis, it guided its development, and it was also a good way to ensure that I liked the environment before starting a thesis, which is still a contract that lasts at least three years.
I would say that the main stages of my thesis can be summarized by a period of floating where we do not know where we are going, then the discovery of something potentially interesting and finally the rush of writing the article to submit it before the deadline.
About my Academic background
I did my high school in my hometown of Pau, where I have done a scientific field. I then decided to do a scientific elementary school at the “Lycée Louis Barthou”, still in Pau, especially because I wanted to continue making music and it would have been complicated to leave (I play the harp, a particularly transportable instrument).
Then I was admitted to the Ecole des Mines in Paris. Contrary to the elementary school that I had enjoyed a lot, I had a much harder time with this part of my curriculum, and when I left school, I felt like I was not able to do anything. At the same time, my sister was doing a master’s degree in cognitive science and telling me what she saw in class, and it seemed fascinating! That’s why after my engineering school I continued with a Master 2 in biomedical engineering, specialty neuroscience. The goal of the master’s was also to target during the final 6 months internship a laboratory that could then host me in my thesis, and this is what happened.
To do research, you have to be comfortable with the subjects related to your field! For me, it is mathematics, computer science, and to a lesser extent the fields to which I apply them (life sciences, physics). But it is possible to conduct research studies in any field, such as literature, economics, philosophy, history… In fact, you just have to have an attraction for the subject related to the domain.
The hardest part of my journey was the time I was in engineering school. Aside from the fact that I was not interested in many courses, it is a very closed environment and I felt that my space only extended from my room to the classrooms. I also felt more like learning to pretend than consolidating real knowledge. This aspect is a defect sometimes criticized by schools that want to be generalists. Fortunately, it was very different in Master 2!
When we talk about a researcher (and much more rarely, a woman researcher) we see someone in a white coat in a closed laboratory, and we identify it to certain subjects only (physics or chemistry in particular). But in fact, there is research in all fields! So some people spend their day in the mountains doing surveys, others interviewing people, or others like me on a computer without touching a test tube.
In the area of artificial intelligence, I think there are several vocations because it’s an interdisciplinary area. You can do it theoretically, and in that case, it will be doing math. But also from a practical point of view, it can be combined with any field.
An important advantage of research is freedom, both in terms of my schedule: I am quite free in my schedules and in the subjects I address: I do in large part what I like.
Not only research but also Hobbies
I’ve been playing harp for several years now, and I’ve always managed to enroll in a conservatory to do chamber music lessons with other musicians.
I also play board games, role-playing games, murder parties and sometimes video games. During the lockdown I discovered crochet and since then I’ve made plenty of stuffed animals (like the one on the right) for my family, and also for my colleagues.
I also go back to yoga and I am very happy to be in the Netherlands because it allows me to go to work by bike!
The final word
It is often thought that to do research you have to be good in all subjects but you just have to be comfortable with the subjects related to your field.
I was sometimes considered fragile or sensitive, especially because of health problems I had in high school. When I’m about to give up, I stand up because there’s no way I’m going to hear again that it’s too hard for me!
Written by Maeva T.