Scientific basis
Cardiac biological age refers to the functional condition of the heart, which may differ from a person’s chronological age—the number of years they have lived. Biological age is influenced by factors such as lifestyle, genetics, and cardiovascular health. A heart that functions poorly due to risk factors like hypertension, obesity, or smoking may have a biological age higher than its chronological age, increasing the likelihood of adverse cardiovascular events. Conversely, a well-maintained heart can have a biological age lower than its chronological counterpart, indicating better health and reduced risk of mortality. The assessment of cardiac biological age provides a more precise measure of heart health and enables targeted interventions for individuals at higher risk.
Techniques and skills
Cardiac statistical shape models provide a comprehensive and nuanced understanding of heart structure. While conventional measurements quantify size, shape models capture variations in anatomical geometry, enabling detailed assessments of structural differences that may influence cardiac function. These models leverage large-scale imaging data and machine learning to identify subtle morphological patterns associated with disease progression, offering insights beyond simple size-based metrics.
Aim
To provide an unprecedented detailed and comprehensive 3D description of how cardiac anatomy ages, and with it of the cardiac anatomy biological age, that demonstrates an improved ability to predict clinical outcomes and mortality. To this ends, the project will involve the construction of 3D anatomical twins of cardiac anatomy, the study of how they change across chronological age in the largest cohort in the world (UK BioBank), the definition of the age gap between the morphology phenotype and the chronological age, the study of factors that contribute to this age gap, and the analysis of the ability of this metric to predict outcomes.
Specific objectives
1. To build a statistical shape model (SSM) of the bi-ventricular anatomy of the heart of UK BioBank participants (n ~85,000) using an existing fully automatic pipeline that has already built ~45,000 models. Improve pipeline robustness and accuracy. (Year 1)
2. To measure biological age of cardiac anatomy by a statistical regression between the anatomical descriptors (enhancing conventional metrics such as mass or volume with SSM derived biomarkers) and chronological age (First half of year 2).
3. To build and validate an ageing model from the two previous steps by studying its plausibility, its association with age-dependent traits, and by its ability to predict the onset of heart failure in UK BioBank and other cardiovascular clinical outcomes (second half of year 2)
4. To interrogate plasma proteomic and metabolomic datasets to uncover molecular signatures associated with divergent cardiovascular ageing trajectories. The so identified readouts will then be systematically mapped to ageing-related biological pathways, mechanisms of cardiovascular resilience, and early-stage pathophysiological processes. This approach will offer mechanistic insights into the biological basis of cardiovascular ageing. (Year 3)
The fourth year of the PhD will then be co-designed with the student, depending on progress and findings, exploring the directions of expansion into (a) a more comprehensive morphological phenotype, e.g. by 4 chamber cardiac models or capturing 3D contraction patterns; (b) mechanistic simulations to investigate hypothesis generated in objective (4) and (c) study the speed of aging using the longitudinal data of UK BioBank.
3 month rotation: objective 1 with the existing 45,000 models.
