Ainhoa García-Sánchez (Spain)
ACE Alzheimer Center Barcelona
Research
I hold a BSc in Physics from Sorbonne University in Paris. Then, after graduating in 2020, I enrolled in a MSc in Management and Analysis of Big Data in Miguel de Cervantes European University. It was during my MSc internship period that I started working in Ace Alzheimer Center Barcelona studying subjective cognitive decline, one of the earliest noticeable symptoms of Alzheimer’s disease and related dementias. The studies were carried out with diverse techniques of machine learning and that was when I realised that I would like to use new technologies such as machine learning and artificial intelligence to improve the field of diagnosis and treatment in the healthcare system. Currently I am working as a bioinformatician at Ace and I am studying for an MSc in Artificial Intelligence in International University of La Rioja.

Presenter of 1 Presentation

 Conference Calendar

REGIONAL NEUROANATOMICAL SUBTYPES IN SUBJECTIVE COGNITIVE DECLINE: A CLUSTERING ANALYSIS IN THE FACEHBI COHORT

Session Name
1760 - AD HETEROGEINITY , DIAGNOSTIC AIDS AND CLINICAL TRIALS (ID 193)
Session Type
SYMPOSIUM
Date
Sat, 19.03.2022
Session Time
05:15 PM - 07:45 PM
Room
ONSITE: 113
Lecture Time
05:45 PM - 06:00 PM
Presenter

Abstract

Aims

Subjective cognitive decline (SCD) is associated with increased risk of mild cognitive impairment (MCI) and dementia such as Alzheimer’s disease (AD). Our aim was to determine differential neuroanatomic subtypes that might serve as a predictive biomarker of cognitive decline using a data-driven approach.

Methods

147 individuals with SCD from the FACEHBI cohort who underwent a 1.5T brain MRI at baseline visit were included. Cortical thickness from 68 regions was estimated using Freesurfer 6.0.1. After reducing dimensionality using a variance filter, Ward's hierarchical agglomerative analysis was conducted to cluster individuals based on their cortical thickness. Differences in clinical variables, APOE-e4 status and Aβ deposition assessed using FBB-PET were examined between subtypes.

Results

Three neuroanatomical subtypes were identified. The first subtype showed increased cortical thickness (non-atrophy subtype, n=49) in comparison to the others. Between these two subtypes, one showed reduced thickness in the lateral temporal (temporal subtype, n=47) and the other in the cingulate/parahippocampal cortex (cingulate subtype, n=51). Subjects in these two subtypes were significantly older and had worse executive function performance. The cingulate subtype also showed a trend towards higher rates of conversion to MCI during follow-up. No significant differences were found for the other variables examined.

Conclusions

SCD individuals showed three distinct cortical subtypes: a non-atrophy subtype and a temporal and cingulate subtypes characterized by older age and worse cognitive performance. The cingulate subtype also showed higher rates of conversion to MCI. These findings may contribute to reveal the earliest neurodegenerative processes underlying AD and to identify those at risk of conversion.

Hide