ScreenIA
The ScreenIA project aims to develop and validate a clinical decision support system for the early detection of preclinical stages of dementia using artificial intelligence techniques. The proposal is based on the integration of digital cognitive markers and voice biomarkers (semantic and paralinguistic) as accessible, non-invasive, and scalable tools for identifying subtle changes before the onset of clinical symptoms.
The project adopts a multimodal and interdisciplinary approach, combining behavioural, cognitive, and technological data with machine learning and deep learning models to improve the sensitivity and specificity of early screening. It includes the development of an integrated predictive model, the experimental validation of the system, and the assessment of its applicability in both clinical and community settings.
The expected impact is to enable earlier diagnoses, optimize healthcare resources, and promote preventive strategies for at-risk populations, thereby contributing to more efficient and personalized care in the context of population ageing.
Key Elements of Ace’s Participation
Technology development and contribution: Ace contributes a voice biomarker-based system, derived from previous research, which is integrated with the digital cognitive assessment platform to build the project’s multimodal predictive model.
Clinical and scientific validation: Ace participates in validating the digital biomarkers and evaluating the model’s effectiveness and accuracy in detecting preclinical stages of dementia, ensuring the clinical relevance and interpretability of the results.
Consortium participation and project implementation: Ace is one of the main partners in the consortium (together with IDES, INTRAS, and UNED) and is involved in project management, data integration, interoperability, and dissemination activities throughout the project’s 36-month duration.
