This study aims to estimate blood pressure levels using data collected from wearable devices, combined with health-related variables from the UK Biobank. With the increasing popularity of smart wristbands and their potential for real-time health monitoring, we intend to develop machine learning models that can predict systolic and diastolic blood pressure based on various physical, lifestyle, and biometric data points.
Can we build accurate predictive models for blood pressure using non-invasive wearable data?
How do demographic, lifestyle, and biometric variables influence blood pressure levels in the general population?
Objectives:
1. Integrate smart wristband-derived variables (e.g., heart rate, activity) with UK Biobank data.
2. Build and validate machine learning models (e.g., random forest, gradient boosting) to predict blood pressure.
3. Identify key features that influence blood pressure levels to inform preventive strategies.
Scientific Rationale:
High blood pressure is a leading risk factor for cardiovascular diseases. Early detection and monitoring are critical. By leveraging large-scale population data and wearable technology, this study contributes to the development of proactive, personalized health monitoring systems that can assist both individuals and clinicians in managing blood pressure effectively.