< Back to previous page


Integrative computer modeling of cardiorespiratory fitness for personalised profiling and prevention of early-stage heart failure

 The surging burden of heart diseases on our society calls for tools that can adequately detect and control heart diseases before disease symptoms present. Clinical exercise tests could improve the characterisation and management of subclinical stages of cardiac disease. Yet, current practice only considers a limited selection of cardiopulmonary exercise indexes in isolation. To utilise the full value of clinical and exercise testing data, in this project, we will apply advanced machine learning (ML) approaches on data collected within the general population (n=450) and in patients with stable coronary artery disease (n≈100). We will develop integrative models that characterise subclinical stages of heart failure and that predict the effect of a 4-month exercise programme on physical fitness and cardiac function. In addition, we will combine innovative immunemetabolic profiling measured by an advanced -omics technique with integrative ML approaches to elucidate key pathways of inflammatory and metabolic stress associated with subclinical heart dysfunction and the cardiopulmonary response to exercise therapy. This pioneering project will lead to novel cardiopulmonary exercise-based algorithms that enable more precise evaluation of cardiac health and boost the development of personalised exercise programmes. As such, these innovative models will enhance the early detection and management of heart diseases in order to counter its epidemic burden on the community.

Date:1 Oct 2020 →  Today
Keywords:heart failure, personalised detection, personalised prevention, Clinical exercise testing, Integrative computer modelling (machine learning)
Disciplines:Biostatistics, Diagnostic radiology, Epidemiology, Biomarker discovery, Cardiology