AI4HealthyAgeing
AI-based Diagnosis Systems for Healthy Ageing
Can we combine EEG with AI models to diagnose ageing related diseases?
Expediente MIA.2021.M02.0007. Programa Misiones de I+D en Inteligencia Artificial.
​
Proyecto subvencionado por CDTI a través de la convocatoria de ayudas Programa Misiones de I+D en Inteligencia Artificial del Ministerio de Ciencia e Innovación del año 2021 y cofinanciado con fondos europeos del Mecanismo de Recuperación y Resiliencia (No. Expediente MIA.2021.M02.0007).
AI4HealthyAging is a project aimed at the early detection of age-related diseases through Artificial Intelligence. It is being implemented by a public-private consortium of 15 entities and has a budget of €12.5 million, funded under the 2021 Artificial Intelligence R&D Misiones program of the Spanish government, Ministerio de Asuntos Económicos y Transformación Digital (Nr. MIA.2021.M02.0007), with funding from the Mecanismo de Recuperación y Resiliencia (MRR).
​
As part of the AI4HealthyAgeing project, Starlab is developing cutting-edge AI technologies to support early detection and monitoring of cognitive decline and Alzheimer’s disease. Our work focuses on the use of electroencephalography (EEG), a non-invasive, low-cost, and scalable technology, combined with advanced machine learning and deep learning algorithms.
​
We are exploring innovative approaches including:
-
Autoencoders for EEG artifact removal and anomaly detection, improving signal quality and enabling more accurate cognitive state assessments.
-
Classification models (Random Forest, XGBoost, SVM) trained on EEG-derived biomarkers to stratify participants by cognitive status (healthy, MCI, AD).
-
Hyperparameter search optimization using cloud computing with AWS, improving the performance of deep neural networks for EEG representation learning.
-
Graph autoencoders for learning brain connectivity patterns and uncovering hidden structures in EEG functional connectivity networks.
In collaboration with Fundación ACE, a leading institution in Alzheimer research and diagnosis, we are integrating high-quality clinical data — including the NBACE neuropsychological battery — and exploring how EEG-based methods can be translated into real-world clinical tools.
​
Starlab’s work in this project contributes to the broader goal of creating accessible, scalable, and AI-powered tools that support healthy brain ageing, early intervention, and more personalized healthcare.
