| Title: | Computational and machine learning tools for the analysis of genetic and medical imaging data to study human health and disease |
| Lead Institution: | Helmholtz Zentrum Munchen |
| Principal investigator: | Dr Francesco Paolo Casale |
Application 87065
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
The availability of medical imaging data in large genetic cohorts enables the study of human health and disease at an unprecedented scale and resolution. For example, these datasets can help identify new quantitative imaging biomarkers of disease, characterise the associated pathological processes and identify their driving genes and pathways through genetic analyses. Despite their great promises, joint analyses of genetic and medical imaging data still present several computational and interpretation challenges and general frameworks for these analyses are not fully established. In this project, we aim to develop computational tools to extract complex phenotypic patterns from medical imaging data, identify their genetic and environmental drivers, and establish their link with human health and disease. To do so, we will extend and combine tools from the fields of deep learning, statistical genetics and causal inference. We propose to use the UK Biobank resource to benchmark these tools across different imaging modalities and indications, including but not limited to brain MRIs for degenerative and neurological disorders, abdominal MRI for metabolic disorders, cardiac MRIs for different cardiovascular diseases and conditions, and DXAs for osteoarthritis. Considering that our project involves the analysis of multiple imaging modalities and the development of general computational tools, we expect it to take several years. Nevertheless, a successful outcome of this project can lead to new disease biomarkers usable in the clinic and insights that could lead to the development of new therapeutics.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 13571 | Genetics-driven risk predictions leveraging the Mendelian randomization framework | Daniel Sens (+5) | 2024 | Genome Research |
Enabling scientific discoveries that improve human health