Saltar para o conteúdo principal

FCCNFCT Digital Services

Access the national information on scientific and technological funding in Portugal

Activity Funded

FRAIL: Intelligent decision support system for FRAILty assessment in the elderly population

FRAIL: Sistema inteligente de apoio à decisão para avaliação da fragilidade na população idosa

Reference
2024.07266.IACDC
Project Start Date
2025-03-01
Project End Date
2026-01-31
Principal Investigator
Scientific Area
Exact sciences
Funding Program
Inteligência Artificial, Ciência dos Dados e Cibersegurança de relevância na Administração Pública

Abstract

The pace of population ageing is increasing worldwide, and current evidence shows that a significant proportion of community-dwelling elder people are frail [1]. Frailty is a geriatric syndrome that represents a status of extreme vulnerability to endogenous and exogenous stressors, exposing the individual to a higher risk of premature death and negativehealth-related outcomes that ultimately imposes a significant burden for the person (his or her family) and public health systems [2]. In Portugal, the mean prevalence of frailty among the community-dwelling population aged 65 years and older was estimated to be ~22%, which is higher than in other countries [3]. The prevalence of frailty rises to>40% among those with chronic diseases. To facilitate risk stratification of frailty in patients, classical scoring systems based on primary care settings and secondary care data such as eFI and HFRS scores have been developed[4, 8]. But it is becoming clear that they are not ready for clinical practice. In addition, these score systems do not consider individual patient characteristics failing to provide more accurate information to guide clear prognostic and screening [9]. Some of these scoresare only appropriated for scientific purposes [10]. The increasing amount of healthcare data that can be accessed in this field, highlights the importance to use machine learning (ML) models and subspace clustering approaches for discovering patterns that can discriminate frailty conditions at earlier stages and be indiciary of actionable changes in patient condition. Nevertheless, although the integration of clinical data has been widely referred as a promising approach for unraveling relevant knowledge for early frailty assessment, no thorough work exists in this domain specific for the Portuguese population. Under this scope, the FRAIL project aims at developing a decision support system that combines intelligent modeling and health data consolidation of elderly patients to improve the accuracy of frailty detection and those at risk of frailty-related negative health outcomes. The expected output is the creation of an innovative screening system based on intelligent modeling, which will have the capability to predict and classify older patients as “high/low” frailty risk (i.e.,who will be at increased risk of frailty) with strict explainability guarantees, and heightened sensitivity to the singularities of the Portuguese population. For that purpose, the computational approaches will integrate heterogeneous groups of health variables – biological profile, sociodemographic characteristics,clinical data, among others. Given the heterogeneous and high-dimensional data nature, the descriptive and predictive modeling will be performed using integrative associative models on subspaces and ML. The models will be extended with principles to handle the structural sparsity of the data, and a solid statistical frame will be pursued to guarantee the statistical significance of the discovered relationships. The resulting decision support system will be offered to the municipality of Famalicão which has physical activity training programs to improve health outcomes in senior persons.

Institutions

Main Institutions

  • Universidade Nova de Lisboa Associação para a Inovação e Desenvolvimento da FCT (NOVA.ID.FCT)

Other Institutions

  • Universidade do Porto Faculdade de Desporto (FADEUP)
  • Câmara Municipal de Vila Nova de Famalicão (MVNF)
  • Instituto de Engenharia de Sistemas e Computadores Investigação e Desenvolvimento em Lisboa (INESC-ID)

Funding 75.435,05 €

Fundação para a Ciência e a Tecnologia (FCT) - Portugal

0,00 €

União Europeia - Estrutura de Missão Recuperar Portugal (UE - EMRP)

75.435,05 €
  • 2025-03-01 a 2026-01-31
  • 2024.07266.IACDC