Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Harnessing the Power of Language: Advancing the Prediction of Individual Trajectories in Aging with an Advanced Cerebro-Cognitive Age Estimation Algorithm Targeting Language Networks
Poster A110 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Elise ROGER1,2, Ana Inés Ansaldo1,2, Simon Duchesne3, Yves Joanette1,2; 1Centre de recherche de l'Institut universitaire de gériatrie de Montréal (CRIUGM) , QC, 2Département d’Orthophonie et d’Audiologie, Faculté de médecine, Université de Montréal, QC, 3Département de radiologie et de médecine nucléaire, Faculté de médecine, Université Laval, QC
It is now possible to characterize an individual's condition on the basis of his/her "cerebral age," derived from observable brain characteristics through neuroimaging and advanced artificial intelligence techniques. Considering the individual's "biological age" (as opposed to chronological age) has emerged as a pivotal approach to identifying health indicators and patterns of age-related diseases. Nevertheless, the conventional measure of cerebral age remains limited as it relies on an overall assessment of brain structure and a single MRI scan. In the PRESAGE (PREdictions and Stimulations related to AGE) research project, we aim to address the limitations of brain age in order to improve our ability to identify individual trajectories of brain and cognitive change. We propose a new method for estimating "cerebro-cognitive" ages which are specific to cognitive systems, and in particular to the functional networks of language. Our preliminary study conducted on test (MTL local cohort, n = 151) and validation (CamCAN, n = 652) databases, has indeed strikingly revealed that in individuals aged 60 and above, brain age predominantly correlates with language-related brain regions and active engagement in cognitive stimulation activities encompassing language and verbal communication. In addition, the literature shows differential impacts on language-related networks (e.g., lexical-phonological, semantic, and control networks) as a function of the form(s) of aging, making language an optimal candidate to significantly improve the diagnostic sensitivity of the measure. Our estimates will be made from a combination of existing databases, encompassing neuroimaging and language data from various adult populations across a wide age range (local databases, CIMA-Q, CCNA/CCNV, UK BioBank). These databases have been selected to ensure adequate representation of different populations, both clinical and non-clinical, and for the development of robust statistical models. The functional validity of the specific brain ages will be assessed by establishing the links with individual cognitive efficiency in several language domains. Finally, we will leverage follow-up scans and evaluations to analyze age-specific derivative functions. These longitudinal ages will allow us to capture the velocity and acceleration of neurocognitive aging across diverse populations, to accurately delineate individual evolutionary trajectories. Overall, the PRESAGE project should actively contribute to improving diagnostic reliability, detecting high-risk profiles, and promoting precision neuropsychology for tailored interventions. From a fundamental perspective, the project should yield new insights and valuable information on the aging patterns of language systems under different conditions.
Topic Areas: Methods, Computational Approaches