Artificial Intelligence in Biomedical Signal Processing

The research department “Artificial Intelligence in Biomedical Signal Processing” (BioSP) develops new signal processing methods for medical technology, including the analysis of biosignals measured on humans, the analysis and processing of acoustic signals, e.g., for hearing aids, and the signal analysis for medical imaging. Methods of artificial intelligence are used to learn complex relationships in a data-driven fashion for which available mathematical models are too limited. Robustness against distortions and incomplete data, as well as interpretability, are of paramount importance.

Research and development areas include:

• Robust and computationally efficient multitask networks
• Transfer learning
• Biosignal analysis for human-machine interaction (e.g., intuitive prosthesis control, rehabilitation tools)
• Acoustic signal processing for hearing aids
• Acoustic event detection and scene analysis
• Sleep staging
• Medical imaging (magnetic resonance tomography, magnetic particle imaging)

The research area of Artificial Intelligence in Biomedical Signal Processing builds upon the fundamental research work at the Institute for Signal Processing, The Universität zu Lübeck, directed by Prof. Dr.-Ing. Alfred Mertins.