黑料福利网

Monomorphic scar-related ventricular tachycardia (VT) is a life-threatening heart rhythm disorder that can occur years after a myocardial infarction (heart attack). Although many patients develop scar tissue after a heart attack, only a small proportion eventually experience VT. Understanding why this occurs, and whether at-risk patients can be identified earlier, remains an important clinical challenge.

Current risk stratification methods rely primarily on measures such as left ventricular ejection fraction and heart failure symptoms. While useful, these methods do not capture the complex, long-term changes that occur over many years. This project investigates whether routinely collected electronic health record (EHR) data can provide earlier and more personalized insights into VT risk.

The research is part of the COMBAT-VT (COMputational model BAsed decision supporT for patients at risk for sustained Ventricular Tachycardias) study. Within this project, we assembled a retrospective cohort of 2,787 patients who experienced a ST-elevation myocardial infarction (STEMI) and were treated at Catharina Hospital between 1997 and 2023. The dataset includes 105 patients who subsequently developed VT and 2,682 who did not. It contains rich longitudinal clinical data, including approximately 6,000 ECGs, 1.6 million laboratory measurements, 270,000 echocardiographic measurements, 1.1 million vital sign measurements, medication records, diagnostic codes, and hospital admission histories.

Using this dataset, we investigate whether routinely recorded 10-second ECGs contain subtle patterns that can predict VT months or even years before the first arrhythmic event. We further explore whether integrating ECG-derived features with laboratory measurements, echocardiographic parameters, and vital signs improves predictive performance and provides new insights into the progression of the arrhythmogenic substrate.

To address these questions, we apply a range of analytical methods, including statistical modeling and machine learning. Beyond developing predictive models, the project aims to improve our understanding of the biological and clinical processes that precede VT while also evaluating the opportunities and limitations of using longitudinal real-world clinical data for cardiovascular research.

Prof.dr. Lukas Dekker

Prof.dr.ir. Frans van de Vosse

This project is part of the COMBAT-VT project (Project No. 17983), which is funded through the High Tech Systems and Materials research program and partially financed by the Dutch Research Council (NWO). In addition, this project is conducted within the framework of the Eindhoven MedTech Innovation Center (e/MTIC) as part of the PICASSSO project (Reference No. TKI HTSM/20.0022). PICASSSO is funded by Holland High Tech | TKI HTSM through the PPS Allowance Scheme for public-private partnerships.