NEURO IMAGING DATE SCIENCE
The Neuro Imaging Data Science group focuses on the development and application of novel and state-of-the-art neuroimaging data analysis tools. In particular, we are interested in the application of machine and deep learning algorithms to find neural biomarkers for neuroscience research, diagnosis, and therapy.
To decode the mechanisms underlying specific brain functions, we combine information from multimodal neuroimaging data such as from magneto- and electroencephalography (M/EEG), magnetic resonance imaging (MRI) or near-infrared spectroscopy (NIRS). However, one of the challenges when combining such complementary datasets is the different activation profiles acting on different spatial and temporal scales. We believe that machine and deep learning algorithms will prove to be valuable tools to overcome such challenges.
Projects
Revealing whole-brain causality networks during guided visual searching
Here, in a first of its kind study, magnetoencephalography was combined with eye-tracking technology to investigate how interregional interactions in the brain change when engaged in two distinct forms of active vision.
Insights into conflict processing networks using MEG
This study uses an adaptation of the Simon task with magnetoencephalography (MEG) to provide detailed insight into the underlying neural mechanisms of the FPAN, with particular focus on its temporal characteristics and directional interconnections.
Here, in a first of its kind study, magnetoencephalography was combined with eye-tracking technology to investigate how interregional interactions in the brain change when engaged in two distinct forms of active vision.
Effect of zolpidem in the aftermath of traumatic brain injury
Zolpidem is a drug commonly used to treat insomnia, but it has also been shown to have a paradoxical therapeutic affect in various disorders of consciousness, such as traumatic brain injury, dystonia and Parkinson's disease.
Deep learning-based classification
We propose an artifact classification scheme based on a combined deep and convolutional neural network (DCNN) model, to automatically identify cardiac and ocular artifacts from neuromagnetic data, without the need for additional electrocardiogram (ECG) and electrooculogram (EOG) recordings.
Zolpidem is a drug commonly used to treat insomnia, but it has also been shown to have a paradoxical therapeutic affect in various disorders of consciousness, such as traumatic brain injury, dystonia and Parkinson's disease.
Real-time data analysis
In order to decode the spatio-temporal organisation of the human brain in vivo, MEG is an essential tool to study transient changes of electrophysiological processes.
Single trial data analysis
MEG source analysis is usually applied to signal averages revealing the most prominent stereotypic activity. The major drawback is that averaging will not preserve the temporal dynamics of each individual response.
In order to decode the spatio-temporal organisation of the human brain in vivo, MEG is an essential tool to study transient changes of electrophysiological processes.
Group Leader
- Institute of Neurosciences and Medicine (INM)
- Medical Imaging Physics (INM-4)
