| Title: | Gender-specific multi-modal phenotyping of hypertensive disease |
| Lead Institution: | University of Oxford |
| Principal investigator: | Ms Winok Lapidaire |
Application 58244
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
Despite healthcare advances, cardiovascular disease remains a major health problem in the world. To reduce disease burden we may need more personalised treatments and we want to study whether there may be specific factors that can be addressed in women. For example, women can develop hypertension during pregnancy, which could accelerate or alter disease progression. Periods of high blood pressure damages small vessels, which leads to a larger heart, altered brain structure, kidney problems and eye disease. In mid-life, damage to these organs has been associated with worse health outcomes for people, even if they have similar blood pressure levels. While previous studies have looked at individual markers of vascular ageing this approach is limited and we want to capture a wider range and combination of changes across the body to better identify people at high risk of clinical events. In sufficiently large databases such as the UK Biobank, machine learning can capture the differences in a wide range of parameters from different modalities between healthy and hypertensive populations. We will do this for men and women separately, as well as for different levels of systolic and diastolic blood pressure: (1) normal, (2) high diastolic, (3) high diastolic and systolic, (4) high systolic. People can then be classified on a scale from health to disease based on their individual measurements. We will then explore how this model works in other younger populations of women, in particular those who have had hypertensive pregnancies.The work will also make it possible to test hypotheses regarding the consequences of high blood pressure in younger adults who have not yet presented with clinical symptoms. Furthermore, we can test whether these subclinical changes in organ structure predispose to risk factors and vascular events later.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 20786 | Contrastive Machine Learning to Quantify Hypertensive Multiorgan Damage and Identify New Disease Phenotypes: A Multinational Multimodal Study. | Mohanad Alkhodari (+25) | 2026 | Circulation |
Enabling scientific discoveries that improve human health