AI Predicts Heart Health 12 Years Ahead

12 Years Before Heart Disease: AI Predicts Future in UAE

The United Arab Emirates has reached another milestone in healthcare innovation. Researchers at the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) have developed an AI-based system capable of predicting the risk of death from heart disease up to 12 years in advance, based solely on two weeks of blood sugar data. The "GluFormer" model could represent a revolutionary breakthrough in preventive medicine, particularly in a country like the UAE, where obesity and diabetes are significant health problems.

A System That Analyzes Metabolism Like a Film

The artificial intelligence known as GluFormer does not work from a snapshot like traditional blood tests but instead creates a "motion picture" of a person's metabolic processes. The system records blood sugar levels every fifteen minutes, enabling it to detect small variations and patterns that routine, periodic blood tests might not reveal.

During research, the system analyzed more than 10 million data points from 10,812 participants—most of whom were not diabetic. GluFormer was able to identify so-called "risk pathways": how the body behaves in various life situations, such as after meals or during sleep, and categorized participants into different risk groups accordingly.

The contrast was striking. In the lowest risk group, no deaths from heart disease were recorded during the 12-year forecast period. By contrast, nearly 70 percent of deaths occurred in the highest-risk group, despite many having normal results in traditional blood tests.

UAE Healthcare Challenges and the Timeliness Factor

The research's significance is particularly prominent in the UAE, where about 40 percent of adults and children are overweight. This tendency places an increasing burden on the healthcare system and contributes to the spread of chronic diseases, such as cardiovascular problems, diabetes, and cancer.

The appearance of GluFormer, on the other hand, could represent not just a prevention opportunity but the beginning of a new era. The system requires only 10-14 days of continuous blood sugar measurement to provide personalized forecasts for future health conditions.

In terms of predictive capability, GluFormer outperformed the current clinical standard HbA1c test in forecasting prediabetic cases—able to predict 66% of cases.

Monitoring Not Just the Heart, But the Whole Body

The system's uniqueness lies in its ability to give an overview of general metabolic health based on blood sugar fluctuations. This includes not just cardiovascular risks but also factors such as visceral fat volume, kidney and liver functions, and lipid profiles—years before any symptoms manifest.

The system's enhanced version already incorporates dining data into the analysis. This makes it possible to more accurately understand individual eating habits and their impact. Among participants who included dining data, predictive accuracy improved by over 90%.

A Digital Twin of Our Body?

Future goals include creating a so-called "digital twin"—a virtual model simulating the person's metabolism, showing how specific lifestyle choices would affect future health conditions.

This concept is not new, but with GluFormer, it has come significantly closer to realization. Meanwhile, its clinical application is not yet ready for widespread deployment. Currently, most healthcare systems are still based on periodic, snapshot-like examinations rather than continuous data collection.

Nevertheless, the UAE is in a prime position to be among the first countries to apply the new system, as it participated directly in its development. According to leading researchers, the country's healthcare institutions are well equipped to conduct further validation studies and prepare for practical implementation.

On the Brink of a New Era

With the advent of GluFormer, the UAE has once again proven that it leads not only in infrastructure and innovation but can also play a pioneering role in healthcare. Such developments serve not only the long-term health of citizens but can also ease the burden on the healthcare system while helping people recognize dangers in time and change their lifestyles.

AI-Based Healthcare Is No Longer a Distant Sci-Fi—The technology is here, ready to fundamentally transform how we view our health. The only remaining question is: Are we ready for this change?

(The article is based on a release by the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI).)