Can data predict who develops arrhythmias after having stroke?

Which heart patients develop arrhythmias in the years after their infarction? Researchers at the Catharina Hart- en Vaatcentrum (cardiovascular center) and TU Eindhoven are trying to find an answer to this question with the help of artificial intelligence (AI). There is still a lot of uncertainty about arrhythmias after having a stroke, which can be life-threatening.

In the COMBAT-VT project, big data and artificial intelligence are used to develop a patient-specific model. This model should help the cardiologist to determine a patient’s risk of developing cardiac arrhythmias and which treatment is most appropriate, such as an implantable defibrillator (ICD) or catheter ablation. Big data will play an increasingly important role in the medical world. The COMBAT-VT project is taking the first steps in using artificial intelligence in the Catharina Hospital.

Patient-specific models

One unique characteristic of the project is that no specific dataset is compiled. Per patient, the researchers examine all available data with which models can be developed. Think of radiological images, EKGs and measurements and data from the patient file. Finding new combinations of factors should ultimately give the researchers a better picture of the patient, which enables them to advise on the best treatment for each patient.

The researchers believe it is very important that the AI algorithms work in an explainable way and can therefore indicate why a patient has an increased risk for these potentially life-threatening arrhythmias. This is essential for the confidence the patient and doctor have in the model.

New requirements for privacy

Because the COMBAT-VT project is so extensive, a pilot has been started in which the researchers are trying to find answers to a number of important questions. For starters: how do we get enough data in a correct manner? This sets new requirements in the context of privacy and regulations.

Data donorship

The researchers suggest that needing large numbers of patients mean that a new way of patient inclusion is needed. It already takes more than 100 days to include the first 50 patients in this pilot. That is why the researchers are making a case for ‘data donorship’. This means that patients are asked for permission at the start of their treatment to share all their data throughout the process.

The current data infrastructure of hospitals is not designed for this kind of very modern research, in which large numbers of data are traced, organized and stored for research. The researchers have already learned a lot about how to do this as well as possible. These new insights will help us to speed up big data research in the future.

Cardiologist Prof. Dr. Lukas Dekker of the Catharina Hospital.

The research is led by cardiologist prof. dr. Lukas Dekker from the Catharina Hospital and biomedical engineer prof. dr.ir. Frans van de Vosse from Eindhoven University of Technology (TU/e). The Catharina Hart- en Vaatcentrum is the largest heart center in the Netherlands and is specialized in the treatment of cardiac arrhythmias.

The COMBAT-VT project falls under the umbrella of the Eindhoven MedTech Innovation Center (e/MTIC), a large regional partnership in the field of medical technology between TU/e, Philips, the Catharina Hospital, Maxima Medisch Centrum and Kempenhaeghe. A crucial collaboration for this research, which combines input from healthcare with technological developments.

Role of the Qualified Medical Engineer

As a trainee of the post-master program Qualified Medical Engineer (QME), Melissa Niemantsverdriet plays a crucial role in developing the clinical workflow for the COMBAT-VT project. She connects the right people and elements with the ultimate goal of collecting, structuring, analyzing and processing patient data for the models.

This is essential for a large data study like this one. QME trainees are educated across the width of the hospital, from IT to patient monitoring, but also management and legislation and regulations are discussed. During the training there is also a lot of attention for the structured approach of projects. This basis, together with a passion for new medical innovations, is why the QME trainee fits well within a progressive project such as the COMBAT-VT project.

/TU/e Public Release. This material comes from the originating organization/author(s)and may be of a point-in-time nature, edited for clarity, style and length. The views and opinions expressed are those of the author(s).View in full here.