Sites Inria

Version française


S.F. (*) - 6/12/2017

Observing cerebral degeneration before it shows

Interprétation des résultats et du traitement statistique d’images médicales - © Inria / Photo C. Morel

Researchers from the Inria centre in Paris and the Brain and Spine Institute (ICM) have, for the first time, discovered very early and almost invisible cerebral and cognitive alterations in people who it is known will later develop frontotemporal degeneration (FTD) or amyotrophic lateral sclerosis (ALS). They are the co-authors of an article on the subject in the medical journalJAMA Neurologyof 2 December 2017. A significant advance for therapeutic research.

Better known under the names Pick's disease and Charcot's disease, frontotemporal degeneration (FTD) and amyotrophic lateral sclerosis (ALS) are neurodegenerative pathologies linked to the progressive loss of neurons in the frontal and temporal lobes and the motor areas of the brain. In the first case they manifest themselves through behavioural disorders, and in the second through progressive muscular paralysis. No treatment exists to cure them. Once the symptoms show, it is too late: the brain cell destruction process, which began years earlier, is well under way. Therapeutic research therefore focuses more on preventive treatments, capable of slowing the appearance of lesions before the first clinical signs appear.

It is therefore necessary to act when there are not yet any symptoms or visible alterations in medical imaging. From then onwards, how can we know if the molecules tested are effective at a stage where there are not yet any markers of the disease? This is the question Dr Isabelle Le Ber, neurologist at the Brain and Spine Institute (ICM) and Dr Anne Bertrand, neuroradiologist at the Pitié-Salpêtrière hospital in Paris, wanted to address when they called on the services of the Aramis project team at the Inria centre in Paris. This team develops mathematical tools for the study of the brain and its pathologies

Discovery of microscopic neurological lesions

The researchers focused on a particular group of subjects: individuals carrying the C9orf72 genetic mutation. As this mutation is dominant, in every instance it leads to the development of lesions that will, depending on the cases, be responsible for an FTD or ALS. This group, put together by Dr Isabelle Le Ber, is one of the most important in the world concerning this mutation. 

By comparing medical images and cognitive tests of carriers and non-carriers of this C9orf72 gene, the researchers discovered that alterations are already present before the age of 40, i.e. around 10 or even 20 years before the first symptoms. These signs are very discreet and had never been observed. They concern difficulty in carrying out small, precise motor function movements (touching the little finger with the thumb, for example) and microscopic lesions, invisible to the naked eye on an MRI. The fact that the researchers have been able to accurately describe them is due to software developed by the Aramis team, called Clinica.

Clinica, the software program that boosted the discovery

Our software makes it possible to carry out complex processing of medical images within the scope of clinical research studies", Olivier Colliot explains. "Doctors do not always have the required expertise to use sequences of sophisticated algorithms. Clinica helps them to combine the different IT tools necessary." Invaluable help, which can facilitate other image analyses conducted as part of cerebral pathology studies (Alzheimer's disease, epilepsy, etc.). This is why the team has made it available for free on the website

The work that has just been published is part of a wider programme, PrevDemALS*, coordinated by Dr Isabelle Le Ber, from the ICM, and which aims for a better knowledge of the presymptomatic stage of FTDs. Co-financed by the Paris public hospital system Assistance publique-Hôpitaux de Paris (AP-HP) and the French National Research Agency (ANR), they have also benefited from another initiative: the Inria/AP-HP research position, which enables hospital doctors in Paris to devote 50% of their time, for a year, to research they carry out jointly with the institute's teams. "The fact that Dr Anne Bertrand was able to join the team on a part-time basis made it possible to achieve results more quickly", Olivier Colliot confirms.


How do you see your field of research evolving?

Olivier Colliot: Medical image processing is undergoing tremendous change. Soon, our work will no longer consist solely in extracting information in images but incorporating it with other types of data - cognitive, biological, genetic - characterising the diseases, in order to build real digital medicine. We hope to design algorithms capable of identifying pathologies from the presymptomatic stages onwards and to predict their evolution. This could be decisive for the discovery of new treatments. 

Keywords: Equipe-projet Aramis Dégénérescence cérébrale JAMA Neurology Clinica Lésions neurologiques Inria de Paris