AI Risk Score Reveals Hidden Hypertension-Related Organ Damage

Hypertension is a leading driver of cardiovascular, cerebrovascular, and renal morbidity, yet organ injury often remains clinically silent until late. Reliance on blood pressure readings alone can miss patients who are progressing toward adverse events. This gap complicates risk stratification and timing of intervention in routine care. Researchers have now developed an artificial intelligence tool to map multiorgan injury patterns in people with high blood pressure.

HyperScore, an artificial intelligence (AI)-derived score led by the University of Oxford, estimates hypertension-related damage across multiple organs before a major cardiovascular event. The team also identified six distinct disease patterns, termed HyperTrajectories, that describe how organ systems are affected in different individuals. This framework is intended to capture heterogeneous pathways of disease that are not apparent from conventional clinic measurements alone.

To build HyperScore, investigators integrated hundreds of measures into a single model. Inputs reflected the heart, brain, kidneys, vasculature, lungs, liver, and metabolic profiles. The goal was to characterize the systemic burden of hypertension beyond a single blood pressure value.

The study applied machine learning to imaging and clinical data from more than 27,000 UK Biobank participants. External validation was performed in a further 5,500 participants from the U.S.-based Atherosclerosis Risk in Communities study. Individuals with higher HyperScores were more likely to experience future cardiovascular problems even when blood pressure alone did not separate risk. Brain changes detected on magnetic resonance imaging were among the strongest indicators associated with hypertension-related damage.

The work, published in Circulation on June 21, 2026, suggests potential use in earlier identification of people beginning to develop problems that could lead to stroke, heart failure, or kidney disease. The researchers emphasize that the approach is at an early stage and is not yet ready for routine clinical use. Further development and clinical studies are needed before adoption in care pathways.

“High blood pressure affects people very differently. Some individuals develop significant damage to the heart, brain or kidneys even when blood pressure is only mildly elevated, while others appear relatively protected despite longstanding hypertension. Our findings suggest that AI methods may help us move beyond treating hypertension based purely on blood pressure numbers, toward a more personalized understanding of how the disease affects the body,” said Dr. Mohanad Alkhodari, first author of the study and a current visiting researcher at the Radcliffe Department of Medicine's Clinical Cardiovascular Research Facility.

Related Links
University of Oxford's Radcliffe Department of Medicine