Using Machine Learning to Improve Early Diagnosis of Acute Cardiovascular Diseases in Emergency Patients

What this Trial Is About

Researchers are developing clinical decision support tools using machine learning ML to improve the rapid diagnosis of acute life-threatening cardiovascular diseases in emergency department patients who have chest pain or difficulty breathing. The study focuses on acute cardiovascular conditions such as acute myocardial infarction, acute heart failure, pulmonary embolism, and acute aortic syndromes. This research aims to enhance diagnostic accuracy, speed up patient management, and reduce medical errors by integrating established diagnostic variables with ML models. The study involves creating and validating ML-based diagnostic tools that combine clinical assessments, biomarkers like cardiac troponin, B-type natriuretic peptide, and D-dimer, and electrocardiogram ECG data. The intervention includes platform development, data pooling, model building, performance comparison, validation, and integration into electronic health records. These efforts build on previous advances such as the MI3 model, BASEL ECG Score, and CoDE-ACS tool, aiming to refine ML ECG interpretation and broaden the diagnostic scope to multiple cardiovascular diseases. Participants with acute chest pain or dyspnea will be observed throughout the study to collect diagnostic data and validate the ML models over three years. The research team will monitor the development and implementation of clinical decision support tools, comparing ML models with current diagnostic methods. The study will be conducted at the University Hospital Basel and focuses on improving cardiovascular diagnostics through AI-driven decision-making tools.

MAchine Learning to Boost the Early Diagnosis of Acute Cardiovascular Conditions