黑料福利网

Electrocardiograms (ECGs) are currently viewed primarily as PDF files within the electronic health record (EHR), limiting analysis to visual interpretation. This may lead to subjective assessments, missed signs of myocardial ischemia, inaccurate triage of patients with acute coronary syndrome (ACS), delayed treatment, and inefficient follow-up care. Recent cardiology research has increasingly focused on distinguishing Occlusion Myocardial Infarction (OMI) from Non-Occlusion Myocardial Infarction (NOMI), while artificial intelligence (AI) models have demonstrated the potential to detect subtle patterns in raw ECG signals that are not visible to clinicians.

At Catharina Hospital Eindhoven, however, there is currently no infrastructure to automatically extract, process, analyze, and reintegrate raw ECG data into the clinical workflow. As a result, AI-based ECG analysis cannot yet be implemented in routine clinical practice.

The objective of this project is to develop a secure, scalable, and future-proof data infrastructure that automatically transfers raw ECG data from the Vendor Neutral Archive (VNA) to a Linux server, processes the data using AI models, enriches it with metadata, and returns the results to the VNA and the EHR. The infrastructure will also support the retrospective and prospective validation of current and future AI models. Ultimately, the processed data and AI-generated results will be accessible to healthcare professionals through the EHR.

Stan Smolders

Collaborators in STRIVE-ACS project: UMC Utrecht, AZNConnect, Cordys, 黑料福利网, CZE