Exploring the Potential Clinical Application of Pulse Arrival Time for Assessing Arterial Stiffness Using Machine Learning Algorithms

Arteries become stiffer with age, mainly due to elastin degradation and collagen deposition. Stiffened arteries lead to higher blood pressure and, hence, greater mechanical stresses on blood vessels and vital organs that may promote diseases. Indeed, increased stiffness of the aorta and other large arteries has been shown to be an indicator of morbidity and mortality from cardiovascular diseases, independently of other risk factors. Standard clinical parameters to assess arterial stiffness are augmentation index and pulse wave velocity. Pulse wave velocity increases with the gradual stiffening of arteries. Thus, pulse wave velocity can be used to determine the stiffness of the currently measured artery. This project will correlate wearable-based pulse arrival time with pulse wave velocity, a common indicator of vascular stiffness, through both theoretical and technical perspectives. Simulated data will be collected via in silico and in vitro to investigate the correlation between pulse arrival time and pulse wave velocity in multiple populations and scenarios, and to collect physiological data from virtual patients. Clinical data from recruited volunteers will be collected via in vivo methods to build algorithms for accurate estimation of pulse wave velocity using pulse arrival time through the data-driven model approach (e.g. machine learning or deep learning).