| Title: | Statistical machine learning of wearable sensor data to predict disease outcomes |
| Lead Institution: | University of Oxford |
| Principal investigator: | Dr Aiden Doherty |
Application 59070
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
Most common diseases are identified too late. For example, over 50% of major cardiac events are in patients not classified as high-risk using current clinical risk prediction methods. Wearable sensors can continuously, noninvasively, and painlessly measure important markers of health in people's everyday lives. However, much more research is needed to test if these sensors really do help us predict who is at risk of common diseases such as cardiovascular disease, type 2 diabetes, breast cancer, colon cancer, etc. We therefore propose to test if the prediction of future disease can be improved by adding data from wearable sensors to clinical risk prediction models. In particular, we will investigate: 1. Can disease events be predicted from wearable sensors alone? 2. Are wearable sensor measurements causally associated with disease outcomes? 3. Are clinical risk prediction models improved by adding wearable sensor measurements?1 Return
| Return ID | App ID | Description | Archive Date |
|---|---|---|---|
| 2040 | 59070 | biobank ACCELEROMETER ANALYSIS: Extracting meaningful health information from large accelerometer datasets | 5 Oct 2022 |
13 Publications
Enabling scientific discoveries that improve human health