The University of Manchester

Academic Institute

OFHS250191

The aim of this study is to develop artificial intelligence (AI) tools that are specifically designed for people of Asian ethnicity in the UK. These tools will be used to predict the chance that someone will develop cardiovascular disease (CVD), such as heart attack or stroke, within the next two years. Research Question: How well can AI models designed specifically for UK Asian populations predict the risk of cardiovascular disease over a two-year period? The study will look at overall accuracy as well as how well the models perform for different Asian sub-groups, including: South Asian Chinese Other Asian Over the past ten years, Artificial Intelligence (AI) has been widely used to estimate the risk of cardiovascular disease (CVD) across the UK population. These AI models are designed to work for everyone, regardless of ethnic background, and they typically achieve around 75% accuracy in predicting future heart disease. However, these are “one-size-fits-all” models. They do not take into account the important differences between ethnic groups. In the UK, people from Asian ethnic backgrounds make up about 9.3% of the population, making them the second largest ethnic group, after the White population (around 82%). Despite this, Asian groups are under-represented in health data, and they are known to have distinct patterns of cardiovascular risk, such as developing heart disease at younger ages or at lower body weights. Because of these differences, AI models that are specifically designed for Asian populations could better capture their unique risk patterns. This could lead to more accurate and fairer predictions of heart disease risk for Asian communities compared with general models built for the whole population.

Heart disease is one of the leading causes of illness and death in the UK. People from Asian ethnic backgrounds are more likely to develop heart disease at a younger age and often experience worse outcomes. However, the tools currently used to estimate future heart disease risk are mainly based on data from White populations. This means they may not work as well for Asian communities and can underestimate risk. This research aims to make heart disease risk prediction fairer and more accurate for UK Asian people. By developing new AI models using health data from Asian participants, the study will better reflect the real risk patterns seen in these communities. This helps ensure that people are not overlooked because existing tools do not fully represent them. More accurate risk estimates can help doctors identify Asian individuals who are at higher risk earlier, even when standard health checks appear normal. Earlier identification allows people to receive advice, monitoring, and treatment sooner, helping to prevent serious events such as heart attacks and strokes. Overall, this study aims to reduce unfair differences in heart disease outcomes and support a more inclusive healthcare system where advances benefit everyone.

Public Health Research

24/07/2026

24/06/2026

TRE