Real-time fMRI neurofeedback to better identify and treat cognitive impairments in epilepsy.

This project concerns the real-time fMRI data acquisition and the signal quality of this data for neurofeedback, finally aiming to improve cognition. This research is performed in the consortium involving Eindhoven University of Technology, Philips Research and Kempenhaeghe R&D.

With real-time fMRI, the images of our brain activity are acquired, processed, and visualised while a person is inside the MRI scanner, essentially providing a window into the human mind as we think, feel, process information, and make decisions. But how well do we understand this technology and its practical implications, especially for healthcare? As a research field, real-time fMRI is undergoing rapid development to support applications like brain computer interfaces, neuroimaging-based interventions for neuropsychological or -psychiatric patients, adaptive experimental paradigms, and neurofeedback training, the latter two being of increased interest in neuroscientific and healthcare research.

They approach the research topic in three parts: (I) Understanding real-time fMRI data quality, in which they present the complexities of the neuronal signal, noise sources, artefacts, and confounding factors, and develop a comprehensive understanding of real-time fMRI data quality; (II) Hardware and software for real-time fMRI analysis and quality control, where they develop, describe and share innovative hardware and software solutions for improving the quality and reproducibility of the real-time fMRI signal; and (III) Real-time multi-echo fMRI, where they explore novel data acquisition and signal processing methods that lead to sensitivity improvements for real-time fMRI.

They have made all of the developed methods and pipelines publicly available as part of a general MATLAB/Octave based toolbox for real-time and offline fMRI preprocessing and quality control.

Read more about the project here.