ERC Starting Grant for Dr. Alexander von Lühmann

Wearable neurotechnology & multimodal machine learning to analyze brain function

The goal of Dr. Alexander von Lühmann's research group at the Berlin Institute for the Foundations of Learning and Data (BIFOLD) at Technische Universität Berlin is to measure and analyze brain activity in people's everyday environments. For this highly complex research project, Dr. von Lühmann has been awarded an ERC Starting Grant of EUR 1.65 million. "Measuring and linking brain activity with human physiology in the context of everyday situations promises profound new insights into brain function and health. I’m thrilled that we can now embark on this exciting project," von Lühmann said.
However, there is still a long way to go from measuring brain activity in a conventional lab environment to doing so in the complexity of natural settings. "The transition to a valid representation and analysis of human brain function in everyday life will yield groundbreaking scientific discoveries and advances in our understanding of neural development, health, and aging, driving progress in medicine and psychiatry," von Lühmann remarks. In 2022, the 36-year-old scientist returned to TU Berlin after conducting research in Boston and serving as director of R&D at a Berlin medical technology company. At BIFOLD, he founded the IBS Lab (Intelligent Biomedical Sensing). He earned his master’s degree in electrical engineering from the Karlsruhe Institute of Technology (KIT) and completed his PhD with distinction at TU Berlin in 2018 under BIFOLD co-director Professor Dr. Klaus-Robert Müller. In his dissertation, he focused on combining functional near-infrared spectroscopy (fNIRS) and electroencephalography (EEG) with multimodal machine learning-based analysis approaches.

The new platform could have unprecedented impacts on neurotechnology applications and research

The lack of suitable mobile neurotechnology has so far been a significant obstacle to monitoring brain activity in everyday life. "Functional magnetic resonance imaging (fMRI) has greatly advanced our understanding of brain functions and networks, but it is limited to snapshots in artificial lab settings. However, we need to better understand the brain in its natural environment. While EEG is mobile, it cannot be directly linked to the brain networks captured by fMRI," von Lühmann explains.

"Our goal is to develop small, innovative, mobile, and wearable neurotechnology that can continuously measure specific brain network activities in natural environments." Von Lühmann and his team are focusing on high-density diffuse optical tomography (HD-DOT), based on near-infrared light, as a suitable alternative to fMRI. They are developing a unique system-engineering concept that miniaturizes and integrates DOT, EEG, and physiological sensors with multimodal machine learning (ML) to improve the spatiotemporal contrast in mobile brain imaging, despite challenging environmental conditions.

Their concept involves four phases: In the first phase, they will develop the necessary hardware for unobtrusive and continuously wearable brain-body imaging using HD-DOT-EEG. In the second, experimental phase, they plan to collect extensive multimodal data using this new technology. Brain networks will be measured and reproduced while the complexity of environmental and physiological artifacts is systematically increased. Phase three focuses on analysis. Here, multimodal sensor fusion and machine learning will enable robust assessment of brain network activity even in complex everyday environments. In phase four, these new technologies will be validated in a fully natural real-world brain imaging experiment.

"I am convinced that now is the right time to harness the transformative potential of wearable multimodal DOT combined with multimodal ML-based data-driven signal processing methods. If we succeed, this new platform could have unprecedented impacts on neurotechnology applications and research, from everyday neuroscience to new forms of digital health," von Lühmann asserts.