Malene Nørregaard Nielsen - Cross-academy PhD Scholarship 2024
Project summary:
Illuminating the potentials of photoplethysmography for patients with atrial fibrillation using explainable artificial intelligence
How are patient characteristics and management expressed in the photoplethysmography (PPG) signal in patients with atrial fibrillation (AF)?
Many wearables already use PPG to monitor e.g. heart rate, and we know that atrial fibrillation can be detected in a ten-second PPG signal. It is however unknown how patient characteristics, hemodynamical changes in an AF episode and management strategies influence the signal, which we will investigate with the use of artificial intelligence.
Project Title
Illuminating the potentials of photoplethysmography for patients with atrial fibrillation using explainable artificial intelligence
Background
PPG is an optical technology that uses light to detect volumetric changes in the peripheral vasculature, i.e. the increase and decrease in capillary volume. It is widely available in wearables such as smartwatches and provides continuous monitoring of the cardiac rhythm.
Aim
This project aims to investigate the unexplored clinical potentials of PPG as an assessment tool for patients with AF. We aim to investigate (1) the impact of risk factors related to AF on the PPG signal, (2) how AF-related hemodynamic changes are reflected in PPG, and (3) how ablation treatment for AF affects hemodynamic normalization.
Methods
We will develop deep neural networks for the detection and characterisation of hemodynamical patterns related to AF based on PPG recordings and characteristics from three cohorts comprising >6500 patients. Specifically, we will use and develop convolutional neural networks and explainable artificial intelligence methods, which allow for a visual interpretation of the otherwise hidden decision-making of the deep neural networks and graphically illustrates the linkage of PPG to AF.
Malene Nørregaard Nielsen
- MSc
- University of Copenhagen , Department of Biomedical Sciences
- Cross academy grant recipient: DCAcademy & Danish Data Science Academy
Main supervisors:
Jørgen Kanters, Associate Professor, Department of Biomedical Sciences, University of Copenhagen
Co-supervisors:
Dominik Linz, Professor, Department of Biomedical Sciences, University of Copenhagen and Department of Cardiology, Maastricht University Medical center and Cardiovascular Research Institute.
Jonas Isaksen, Postdoc, Department of Biomedical Sciences, University of Copenhagen