Skip to content
TU Dresden
A joint project with Uni Leipzig DE EN

Artificial Intelligence in the Context of Biomedical Signals

Mentored by Jochen Hampe & Jakob Kather
at EKFZ for Digital Health Dresden, TUD and UKD Dresden

The analysis of ECG signals is a key component in the detection and prediction of cardiac disease. In this research topic, we will explore the use of machine learning and neural networks for the prediction of cardiac attacks using ECG signals. The second research topic is the automation of sedation using deep learning analysis of multidimensional biosignals. This research will focus on developing a system for automatic monitoring of sedation levels in patients undergoing medical procedures.

Key research questions for this topic include:

  • Improved detection algorithm for vectorcardiograms and reconstructing diagnostical ECG from orthogonal ECG.
  • How can we incorporate variation in ECG signals due to different physiological factors, such as age and gender, in the synthetic data generation process?
  • How can we leverage prior knowledge from the medical domain, such as anatomical knowledge, to improve the accuracy of the predictive models?
  • How can we evaluate the performance of the models in predicting cardiac attacks and in the automatic monitoring of sedation levels using multidimensional biosignals?

Work Environment

The research will be conducted in close collaboration with medical experts in cardiology at the University Hospital Dresden (UKD) and the EKFZ for Digital Health. The EKFZ for Digital Health offers a unique research platform with a Working Lab in cooperation with AI experts in the field of oncology, which provides an ideal environment for interdisciplinary research in digital health.

Prerequisites

  • Master's Degree (or equivalent) in computer science, electrical engineering, applied mathematics, or related fields of expertise
  • Very good programming skills, with practical experience in Python
  • Strong background and practical experience in time series analysis using PyTorch or equivalent packages
  • Ability to collaborate effectively in an interdisciplinary environment
  • Strong motivation to publish and present research results at conferences.