idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
A landmark international study led by the University of Zurich has shown that artificial intelligence can assess patient risk for the most common type of heart attack more accurately than existing methods. This could enable doctors to guide more personalized treatment decisions for patients.
Doctors caring for patients with the most common form of heart attack – the so-called non-ST-elevation acute coronary syndrome (NSTE-ACS) – have so far relied on a standardized scoring system. Using the GRACE score, they can estimate risk and determine the optimal timing for catheter-based treatment. This score is widely used and increasingly integrated into international clinical guidelines. However, it has long been recognized that existing tools cannot always capture the full complexity of these patients.
Data from over 600,000 patients
A new study published in The Lancet Digital Health now suggests that many patients may need to be re-classified, with important implications for how heart attacks are treated worldwide. In the largest study on risk modeling in NSTE-ACS to date, an international research team led by the University of Zurich (UZH) analyzed health data from more than 600,000 patients across 10 countries.
They used artificial intelligence (AI) to analyze clinical trial data from the landmark VERDICT trial and taught the model to recognize which patients benefit most from early invasive treatment, including angiography and stenting.
New risk classification needed
“The results were striking. While some patients gained substantial benefit from early intervention, others showed little or no benefit,” says first author Florian A. Wenzl from UZH’s Center for Molecular Cardiology, who also conducts research at the National Health Service in the UK. According to the researchers, this indicates that current treatment strategies may in some instances be targeting the wrong patients. A major re-stratification of patient care – one that assesses the individual benefit of established treatment strategies – may therefore be necessary in hospitals worldwide.
According to Wenzl, the study illustrates how AI could transform the treatment of heart attacks: “By re-analyzing clinical trial data, our model GRACE 3.0 learned who actually benefits from early invasive treatment – and who does not. This may imply a shift in how we should be managing these patients.”
More personalized and effective care
Last author Thomas F. Lüscher, who conducts research at the Center for Molecular Cardiology in Zurich and the Royal Brompton and Harefield hospitals in London, explains: “GRACE 3.0 is the most advanced and practical tool yet for treating patients with the most common type of heart attacks.” The new score not only predicts risk more accurately but can also be used to guide more personalized treatment decisions. “This could reshape future clinical guidelines and help to save lives,” Lüscher adds.
The researchers hope that the new GRACE 3.0 score provides doctors in routine clinical practice with a simple, validated and AI-powered tool to deliver more personalized and effective care for heart attack patients.
Dr. med. Florian A. Wenzl
Center for Molecular Cardiology
University of Zurich
florian.wenzl@uzh.ch
+41 76 749 42 50
Prof. Dr. med. Thomas F. Lüscher
Royal Brompton and Harefield hospitals
London, UK
t.luescher@rbht.nhs.uk
+44 7502 008 487
Wenzl FA, Kofoed KF, Simonsson M, Ambler G, van der Sangen NMR, Lampa E et al. Extension of the GRACE score for non-ST-elevation acute coronary syndrome: a development and validation study in ten countries. Lancet Digit Health. 18 October 2025. DOI: https://doi.org/10.1016/j.landig.2025.100907
https://www.news.uzh.ch/en/articles/media/2025/AI-heart-attack.html
Merkmale dieser Pressemitteilung:
Journalisten
Biologie, Informationstechnik, Medizin
überregional
Forschungsergebnisse, Wissenschaftliche Publikationen
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
