A privacy-aware approach to identify individuals at high risk for cardiometabolic diseases using nation-wide registries and genetic information.
Bidragets beskrivning
Cardiometabolic diseases are the leading cause of death in Finland and it is important to early identify individuals at high risk. Traditionally, this is done using clinical prediction models. However, these approaches are limited. I propose to innovate current approaches within a framework that guarantees individual's privacy by: 1) Exploring a new class of artificial intelligence (AI)-based models to generate “synthetic” registry data. This will allow to create new data that can yield the same results of the original data but are different enough to minimize privacy loss. 2)Developing AI-based predictive models that use nationwide registry data on medications, diagnoses, familial risk and social-demographic information to obtain accurate estimates of risk for cardiometabolic disease. 3) Integrating genetic scores for cardiometabolic diseases and health-trajectories to identify subgroups of individuals for which genetic information might improve risk prediction.
Visa merStartår
2019
Slutår
2024
Beviljade finansiering
Andra beslut
352707
Akademiforskarens forskningskostnader(2022)
200 000 €
328891
Akademiforskarens forskningskostnader(2019)
299 999 €
Finansiär
Finlands Akademi
Typ av finansiering
Akademiforskare
Utlysning
Övriga uppgifter
Finansieringsbeslutets nummer
323116
Vetenskapsområden
Data- och informationsvetenskap
Forskningsområden
Laskennallinen tiede
Identifierade teman
public health, occupational health