| Title: | Assessment of frailty and biological age using multi-modal deep learning |
| Lead Institution: | Imperial College London |
| Principal investigator: |
Application 87802
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
Over the next 3 years, we will attempt to develop an artificial intelligence solution that will predict the biological age of a patient's organs. This model will take into account a subject's genetic and lifestyle information and analyse the images of the target organ to make its prediction. A large part of the research will be concerned with how best to combine these disparate data sources for maximum performance. We hold this to be a promising and impactful pursuit due to the effect biological age has on a patient's disease risk and response to medical interventions. It is crucial for doctors to understand how healthy a patient's organs are when considering a diagnosis and also before recommending certain interventions due to the variable risk of complications. We know that lifestyle and genetics are what define our biological state, and thus we expect that combining this information with images of the organ of interest will allow us to make the best and most informed prediction possible. Furthermore, by analysing which parts of the input contributed most to our prediction, we can begin to understand the causal roots of biological-chronological age gaps.2 Publications
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 14566 | Unlocking the diagnostic potential of electrocardiograms through information transfer from cardiac magnetic resonance imaging | Özgün Turgut (+7) | 2025 | Medical Image Analysis |
| 14098 | Using UK Biobank data to establish population-specific atlases from whole body MRI | Sophie Starck (+6) | 2024 | Communications Medicine |
Enabling scientific discoveries that improve human health