Researchers in Australia have developed an artificial intelligence (AI) model designed to analyze DNA and identify genetic risk factors associated with stroke, aiming to enhance personalized risk assessments and preventative strategies. The study focused on ischemic stroke, the most prevalent form of the condition, and identified 18 genetic variants linked to an increased likelihood of experiencing this type of stroke.

According to a statement from the University of Technology Sydney (UTS), individuals classified within the highest genetic risk group exhibited a 22 percent greater chance of suffering an ischemic stroke compared to those in lower-risk categories. The AI model was constructed using a transformer-based approach, a form of machine learning that enabled the system to process thousands of genetic variations and detect their relationship with ischemic stroke.

To validate the model’s effectiveness, researchers tested it against genetic and health data sets from the UK Biobank and a sizable database in the United States. The results indicated that the AI-powered tool outperformed existing genetic risk prediction methods, showing greater accuracy in distinguishing between those who had experienced a stroke and those who had not.

An additional finding of the study highlighted the gene EDEM2, located in brain cortex tissue, as a significant contributor to stroke susceptibility. This discovery offers a potential target for future research into stroke mechanisms and therapeutic interventions.

Building on this progress, the research team is developing a multimodal AI system that integrates genetic data with other sources of information, including health and lifestyle questionnaires, medical imaging, and scans. This comprehensive approach is expected to further refine risk assessments not only for stroke but also for other complex diseases, supporting more tailored prevention and treatment efforts.