The research project aims to develop clinical decision support tools integrating established diagnostic variables and machine learning (ML) models for rapid diagnosis of acute life-threatening cardiovascular conditions in emergency department (ED) patients with chest pain or dyspnea with the ultimate goal of Improved diagnostic accuracy, faster patient management, and reduced medical errors.
None
age < 18 years old
patients presenting in cardiogenic shock
chronic terminal kidney failure requiring dialysis
Dataset of about 200'000 extensively characterized patients enrolled in randomized controlled trials and observational studies are used
Cohort
Other
Machine learning based development of a diagnostic tool for acute cardiovascular disease
Other interventionMALBEC will be delivered through five integrated work packages (WP) encompassing: (0) platform development and implementation, (1) data pooling, (2) model development, (3) performance comparison, (4) performance validation, and (5) platform plugin
Developing a clinical decision support tool
Developing and implementing a clinical decision support tool that integrates and visualizes results of established diagnostic variables in a dashboard
Validate machine learning (ML) models
Derive and validate ML models that integrate cardiac biomarkers with key clinical information and the digital 12-lead ECG to rapidly inform the diagnostic probability for six acute life-threatening cardiovascular conditions in patients presenting with acute chest pain and/or acute dyspnoea to the Emergency Department
Click on the lead sponsor to view all of their trials.
University Hospital, Basel, Switzerland
Lead sponsor
University of Basel
Collaborator
