Prediction of pericardial effusion using a 12-lead electrocardiogram analyzed by artificial intelligence

Our project emphasizes the integration of advanced artificial intelligence techniques with traditional diagnostic methods, fostering a synergy that enhances the accuracy and efficiency of cardiac health assessments. By leveraging the power of Artificial intelligence, specifically through the utilization of a Convolutional Neural Network (CNN), we can discern subtle patterns and nuances in ECG data that may not be readily apparent to human observers. This not only enables us to detect pericardial effusion cases earlier but also enhances our ability to classify them into distinct categories based on severity, such as identifying cases with hemodynamic compromise. Ultimately, this novel approach holds the potential to revolutionize diagnostic capabilities in cardiovascular health, leading to more precise and timely clinical decisions. Through collaboration with participating institutions, our project aims to not only advance medical research but also directly benefit patients by improving the accuracy of diagnosis and treatment in diverse cardiac scenarios.

Faculty Supervisor:

Robert Avram

Student:

Partner:

National High School of Engineering of Tunis

Discipline:

Engineering

Sector:

Artificial Intelligence

University:

Université de Montréal

Program: