Bringing AI to Cardiac Medicine

One heartbeat follows the next on the practice monitor: 24 hours of long-term ECG, around 100,000 recorded heartbeats — and only eight minutes of paid time for the doctor to assess them. Is the patient's heart beating in time, or is it stumbling, racing, and skipping beats? The recording is a helpful tool in detecting cardiac conditions ranging from atrial fibrillation and arrhythmia to other irregularities, and the process can save lives. Yet official statistics reveal that misdiagnoses are more common than many patients realize — on average, around 25 percent.

This is precisely where the "FACE" project, in which the University of Siegen plays a key role, comes in. Together with project partners GETEMED and the Berlin Institute of Health (BIH), the "Medical Informatics and Graph-based Systems" research group led by Professor Kai Hahn and Dr. Christian Weber is developing and assessing AI systems to support doctors in the challenging task of evaluating cardiac data quickly and reliably. The total project budget is just under seven million euros, with the University of Siegen receiving around 430,000 euros in funding.

Die Forschungsgruppe „Medizinische Informatik und graphbasierte Systeme“

The research draws on real patient data: The Siegen research group receives ECG training data free of charge from the world-renowned Massachusetts Institute of Technology — data that has already been expertly evaluated and cleared for research use. The data sets serve as training material for multiple AI systems, with researchers systematically comparing their performance: How quickly do the systems deliver results? How much computing power is needed? And above all, how precise are the ECG analyses?

The best-performing systems are then further developed using internal project data sets. The artificial intelligence is taught to recognize noise and interference, as well as to classify various cardiac arrhythmias, such as atrial fibrillation or atrial flutter, and the normal sinus rhythm — all autonomously, without human assistance.

The project brings the University of Siegen together with five partner organizations, including manufacturers of ECG devices and pacemakers, and the Charité hospital in Berlin. A feasibility study with the practice partners will show how medical professionals envision working with AI and what existing knowledge can be leveraged. One non-negotiable point: The AI model must be data protection-compliant, and patient data must remain secure at all times. Prof. Hahn and Dr. Weber's team is therefore pursuing an edge solution for use in clinics and doctors' practices. In this configuration, the AI reads and analyzes the ECG data directly on site. Sensitive patient data never leaves the clinic or practice, nor is it sent abroad or to third-party providers, as would be the case with cloud solutions on external servers. As such, edge systems are not only much more data protection-friendly, but also run independently of network access.

Edge systems of this kind require proper hardware that can receive ECG data and run the AI analysis. Interviews with practice partners made it clear that for the AI to actually be used in everyday practice, these laptops must be affordable. "Most practices simply can’t afford to spend several thousand euros for the readout hardware, especially if that laptop is to be used exclusively for reading ECGs," Prof. Hahn notes. The research group has accordingly adjusted its mission to include developing a system that works on hardware costing around 1,000 euros. As such, architectural considerations play a central role in the research, with the AI requiring no more computing power than a standard laptop can deliver.

The potential benefits are enormous, for medical professionals and patients alike. AI analysis is already more reliable than human evaluation, detecting abnormalities more precisely and in a fraction of the time. "Of course, AI systems should only support doctors. A human must always check the AI analysis," explains Jasmin Freudenberg, a doctoral researcher from the research group. At present, AI cannot yet derive specific suspected diagnoses from abnormalities. However, Annika Steiger, also a doctoral researcher in the "Medical Informatics and Graph-based Systems" research group, believes that this could become a reality in the medium term.

GETEMED and the BIH are also developing cloud solutions for other target groups. They promise better performance, with the goal of making edge and cloud algorithms available as two distinct alternatives, each harnessing the advantages of its respective approach.

As the "FACE" project is pre-competitive research, the results may not directly lead to a market-ready product. The insights gained will certainly flow into upgrades to devices produced by the participating companies. This might, for example, include a smart ECG device that does not require an additional laptop.

The overall "FACE" project (research into the application of cloud and edge computing for efficient AI-supported ECG analyses) is funded by the former Federal Ministry for Economic Affairs and Climate Protection, with additional sponsorship through the VDI Technology Center, Düsseldorf. The project is scheduled to run from 2024 to 2026, all in service of making cardiac diagnostics more sustainable, precise, and safe.

Cooperation partners: