A study led by a researcher from the Medical University of Gdańsk has been published in Radiology: Artificial Intelligence (Impact Factor >20), one of the highest-impact journals in the field of medical imaging and artificial intelligence. The publication represents a major achievement for the MUG and the University Clinical Centre and is among the most impactful contributions to radiology and neurosurgery research at the institution.
The study introduces a novel artificial intelligence system designed to support the assessment and treatment of brain aneurysms. The technology, developed and patented by the research team, enables rapid and fully automated quantification of aneurysm morphology, including geometric and hemodynamic-related features, which are key predictors increasingly recognised as critical for rupture risk assessment. The system reduces the time required for analysis from approximately 30 minutes of manual work to only a few seconds, enabling faster and more consistent clinical decision-making.
Determining whether an unruptured intracranial aneurysm should be treated or monitored remains one of the most challenging problems in neurovascular medicine. While clinical tools such as the PHASES score are widely used, they rely primarily on demographic and clinical variables and have known limitations in predicting rupture risk accurately. In contrast, aneurysm morphology has been shown to be one of the strongest predictors of rupture, but its use in clinical practice has been limited by the time-intensive nature of manual measurement.
The research was led by Samuel D. Pettersson, 6th-year M.D. doctoral student at the MUG, who served as first author and co-inventor of the technology. The invention was developed in collaboration with a professor from Harvard Medical School and a co-inventor affiliated with Harvard University, as part of a broader international consortium of leading academic and clinical centers.
The study was conducted over three years in collaboration with institutions across Europe, North and South America, including the Medical University of Gdańsk and University Clinical Centre, Beth Israel Deaconess Medical Center at Harvard Medical School, Vienna General Hospital, Hospital das Clínicas of the University of São Paulo, and Buffalo General Hospital in New York.
The article is available here.
