You are here : Home > The Institute > News > Alzheimer's: a more effective method to diagnose the disease?

Scientific result | Alzheimer's disease | MRI

Alzheimer's: a more effective method to diagnose the disease?


A study conducted by Maxime Bertoux (Unit 1171 Inserm / Lille University / CHU Lille) in collaboration with researchers from the Centre Hospitalier Saint Anne, the SHFJ (CEA-Joliot), NeuroSpin (CEA-Joliot) and the ICM shows that the morphological analysis of the cortical sulci would make it possible to recognize Alzheimer's disease in 91% of the cases, against 80% now by anatomical analysis of the cortex.

Published on 20 September 2019

Atrophy of the hippocampus is one of the first signs of Alzheimer's disease. The anatomical analysis of the brain by MRI can detect the disease correctly in about 80% of cases. It usually consists in measuring the thickness of the cerebral cortex (the tissue also called "gray matter", which covers the two hemispheres of the brain) or the volume of several regions of the brain, including hippocampus. According to researchers from Inserm, the University of Paris, the Joliot Institute (NeuroSpin and SHJF) and the ICM, the analysis of cortical sulci morphology by MRI would do better. Their study was published in Neurobiology of Aging.

Sulci are folds of the brain which tend, during aging, to widen. This evolution of the grooves is accompanied by a decrease in the thickness of the cortex which borders them. The same group of researchers has previously highlighted the acceleration of this phenomenon in Alzheimer's disease. The present study aimed to investigate whether the morphological analysis of the brain sulci could make it possible to detect a specific signature of the disease, which would constitute a diagnostic marker of the disease and of its stage of evolution.

The researchers performed brain MRI in 51 patients with Alzheimer's disease including some at an early stage and others at an advanced stage, as well as 29 control participants not affected by the disease. The diagnosis was made after a biological assessment, based on a lumbar puncture for the presence of the biomarkers of the disease and on a positron emission imaging (PET-scan), showing the amyloid deposition characteristic of certain neurodegenerative diseases (accumulations of protein aggregates).

The researchers then used the Morphologist software recently developed at NeuroSpin (CEA-Joliot), which recreates a negative "mold" of the brain from an MRI. The software then extracted in 18 regions of each cerebral hemisphere, an average value of the width of each groove and the thickness of the cortex bordering them. In parallel, the researchers performed the usual measurements of the volume of several brain regions and the thickness of the cortex, in order to compare these techniques.

An algorithm was then used to correlate the measurements to the state of health of each participant (control or patient). The researchers found that the width of a group of few sulci, in particular in frontal and temporal regions, was associated with Alzheimer's disease. It allowed to determine the state of health of the participants in 91% of the cases, against only 80% for the usual anatomical measurements. In addition, the sulcal morphology seems to change with the stages of the disease: they were larger in patients with the most advanced cognitive declines.

"These measures reflect the evolution of the disease and appear to be correlated to the cognitive deficits,” says Maxime Bertoux. “It can be very useful in clinical trials evaluating the effectiveness of a potential drug. In addition, these measures require only an MRI and a largely automated analysis that can be performed in many care centers. This technique still needs to be validated on larger patient cohort, but it could be of great clinical interest," he concludes. The researcher is already using this new approach to detect specific signatures of other neurodegenerative diseases, particularly frontotemporal dementia.

Top page

Top page