My UM
My UM
Search

Advanced signal processing, modeling and machine learning

Assoc. Prof. Stef Zeemering

Research background
With a robust foundation in applied mathematics and computer science, my research has centered around computational cardiac electrophysiology. My academic path began with a Bachelor's degree in Knowledge Engineering from Maastricht University, followed by a Master's degree in Operations Research. After gaining industry experience as a mathematical consultant, I pursued a Ph.D. at Maastricht University, focusing on sparse optimization in mathematical systems theory. This was followed by a position as a scientific software engineer at Maastricht Instruments. In 2011, I joined the Department of Physiology at Maastricht University. My expertise in both commercial and scientific software development has enabled me to create robust tools for analyzing electrical conduction in the atria, tools which are now utilized by various academic partners. My research primarily focuses on measuring and quantifying atrial fibrillation (AF) properties in both animal models and human patients through direct atrial measurements and noninvasive electrocardiograms.

Expertise
My research interests lie in advanced signal processing, modeling, and machine learning techniques applied to AF complexity quantification, mechanism identification, and treatment outcome prediction. I specialize in developing high-quality, innovative tools that integrate data from various sources to analyze the properties and dynamic behavior of a wide range of cardiac signals. These signals include intracardiac electrograms recorded by electro-anatomical mapping systems, high-density epicardial direct contact maps, trans-esophageal catheters, and body surface electrocardiograms. I expanded my expertise to complex genetics, adopting a systems biology approach to understand AF by linking differences in atrial gene expression profiles, determined by next-generation RNA sequencing, to atrial tissue characteristics and patient phenotypes. Although my primary research focus is AF, my expertise is applicable across other areas of cardiovascular research.

Projects
My current projects aim to develop computational tools for the personalized treatment and management of AF patients. Key objectives include:

  • Developing a multi-modal computational framework to quantify key AF characteristics in individual patients.
  • Employing a systems biology approach to understand AF pathophysiological mechanisms and dynamics by integrating data from various biological scales and computer modeling.
  • Translating experimental findings into clinical decision support systems for personalized AF management, risk prediction, progression tracking, and treatment outcome prediction.

A Multi-Modal Computational Framework to Characterize Atrial Fibrillation
I am advancing quantitative methods to assess electrophysiological mechanisms in the atria through:

  • Contact Mapping and Electro-Anatomical Mapping: Identifying candidate driver regions with repetitive conduction patterns during AF.
  • Electrocardiograms (ECG) and ECG Imaging: Noninvasive assessment of AF complexity and atrial remodeling by quantifying the spatiotemporal complexity of atrial activity projected onto the body surface.
  • Transesophageal ECG (TE-ECG): Semi-invasive evaluation of conduction characteristics in the left atrium, often showing pronounced electrical and structural remodeling associated with AF.
  • Implantable Loop Recorders and Rhythm Monitoring Devices: Long-term monitoring of AF dynamics (incidence, duration, and circadian rhythm) to determine AF burden, triggers, and progression.

These methods are complemented by computer modeling of the atria, enhancing the interpretation and quantification of AF patterns, complexity, and driver identification.

Computational Systems Biology of Atrial Fibrillation
I am implementing a systems biology approach to understanding AF by integrating genetic, molecular, and cellular atrial tissue characteristics with electrophysiological measurements. This work involves a detailed 3D anatomical model of the atria, developed in collaboration with the Institute of Computational Science in Lugano. The model incorporates electrical and structural alterations in the atria to simulate their effects on conduction patterns and their body surface projections.

Decision Support for Atrial Fibrillation
My research translates experimental findings into clinical decision support systems, including: 

  • ECG-Enriched Prediction Models: For AF risk, progression, and treatment outcome.
  • AF Driver Identification: Utilizing advanced spatiotemporal pattern detection in novel multi-polar catheters within electro-anatomical mapping systems.
  • Machine Learning & Deep Learning Techniques: To optimize feature extraction from ECGs for treatment outcome prediction.

My focus remains on addressing clinically relevant research questions and developing robust, validated methodologies and user-friendly tools that facilitate the translation of basic research findings into clinical practice. I continuously seek new opportunities to extend the impact of my tools to other cardiovascular applications.

Key publications

  • Gonçalves Marques, V., Gharaviri2, A., Özgül, O., Pezzuto, S., Auricchio, A., Bonizzi, P., Zeemering, S., & Schotten, U. (2024). A novel sequential endocardial mapping strategy for locating atrial fibrillation sources based on repetitive conduction patterns: An in-silico study. Journal of Molecular and Cellular Cardiology Plus, 7. https://www.sciencedirect.com/science/article/pii/S2772976124000059
  • Ozgul, O., Hermans, B. JM., van Hunnik, A., Verheule, S., Schotten, U., Bonizzi, P., & Zeemering, S. (2023). High-density and high coverage composite mapping of repetitive atrial activation patterns. Computers in Biology and Medicine, 159(1), Article 106920. https://doi.org/10.1016/j.compbiomed.2023.106920
  • Winters, J., Isaacs, A., Zeemering, S., Kawczynski, M., Maesen, B., Maessen, J., Bidar, E., Boukens, B., Hermans, B., van Hunnik, A., Casadei, B., Fabritz, L., Chua, W., Sommerfeld, L., Guasch, E., Mont, L., Batlle, M., Hatem, S., Kirchhof, P., ... Schotten, U. (2023). Heart Failure, Female Sex, and Atrial Fibrillation Are the Main Drivers of Human Atrial Cardiomyopathy: Results From the CATCH ME Consortium. Journal of the American Heart Association, 12(22), Article e031220. https://doi.org/10.1161/JAHA.123.031220
  • Zeemering, S., Isaacs, A., Winters, J., Maesen, B., Bidar, E., Dimopoulou, C., Guasch, E., Batlle, M., Haase, D., Hatem, S. N., Kara, M., Kääb, S., Mont, L., Sinner, M. F., Wakili, R., Maessen, J., Crijns, H. J. G. M., Fabritz, L., Kirchhof, P., ... Schotten, U. (2022). Atrial fibrillation in the presence and absence of heart failure enhances expression of genes involved in cardiomyocyte structure, conduction properties, fibrosis, inflammation, and endothelial dysfunction. Heart Rhythm, 19(12), 2115-2124. https://doi.org/10.1016/j.hrthm.2022.08.019
  • Maesen, B., Verheule, S., Zeemering, S., La Meir, M., Nijs, J., Lumeij, S., Lau, D. H., Granier, M., Crijns, H. J., Maessen, J. G., Dhein, S., & Schotten, U. (2022). Endomysial fibrosis, rather than overall connective tissue content, is the main determinant of conduction disturbances in human atrial fibrillation. EP Europace, 24(6), 1015-1024. https://doi.org/10.1093/europace/euac026