Identification and prediction of risk potentials of elderly through digital biomarkers
Successful treatment of the elderly requires the consideration of both the individual disease and the accompanying multimorbidity. The associated risk potentials are determined e.g. through so-called comprehensive geriatric assessments. Risk potentials in the elderly that are discovered too late lead to significant additional costs and poorer patient outcomes. The geriatric expertise needed for risk stratification is usually not available at all relevant key points in healthcare. Furthermore, the information necessary for risk stratification is often not available in a structured way.
To identify risk potentials at an early stage, JOANNEUM RESEARCH HEALTH, Predicting Health, Kärntner Landeskrankenanstalten-Betriebsgesellschaft (KABEG) and Steiermärkische Krankenanstaltengesellschaft (KAGes) are working on an innovative approach for risk stratification. The aim of the project is to develop a digital method for resource-efficient, (partially) automated identification or prediction of risk potential in the elderly using digital biomarkers. In the future, this method should be available for widespread use at key points in healthcare. Artificial intelligence (AI) methods will be used to identify risk potentials such as frailty, delirium (acute confusional state), dysphagia (difficulty in swallowing) and tendency to fall.
The conceptual design and development is carried out in an interdisciplinary manner together with future users within a co-creation process. This process includes interdisciplinary focus groups and interviews, as well as process observations and workshops.
- Duration: 2022 - 2023
- Status: under development
- Funding: FFG
- Cooperation with the association QiGG (Qualität in der Geriatrie und Gerontologie), KABEG, KAGes and Predicting Health
- Gutheil, J; Donsa, K; (2022) SAINTENS: Self-Attention and Intersample Attention Transformer for Digital Biomarker Development Using Tabular Healthcare Real World Data. DOI: 10.3233/SHTI220371