Fast functional magnetic resonance imaging: technical developments and applications

Professor Lin Fa-Hsuan will talk about how the temporal sampling rate of fMRI can be improved while maintaining a good spatial resolution.

Functional magnetic resonance imaging (fMRI) using blood-oxygen-level-dependent contrast has become a dispensable tool in non-invasive human brain function studies. Conventionally, fMRI experiments use echo-planar imaging (EPI) as the data acquisition method. By using fast switching of the imaging gradients, EPI can offer whole-brain coverage, isotropic spatial resolution of about 3 mm, and 0.5 Hz sampling rate. While fruitful results have been obtained by fMRI with this spatiotemporal resolution, the quest of even faster fMRI data acquisition is still desired in order to further improve the sensitivity of detecting hemodynamic responses elicited by tasks and cognitive processes.

Enabled by highly parallel detection of the fMRI signal, the inverse imaging  (InI) method developed in our lab can push the fMRI sampling rate up to 10 Hz using a 32-channel head coil array at 3T with approximately 5 mm spatial resolution at cortex. In this talk, we will summarize technical developments of InI. Importantly, we will also present applications of this fast fMRI method in monitoring/suppressing physiological noise and detecting fine temporal information in hemodynamic responses. Supported by preliminary results, we hypothesize that BOLD fMRI signal carries physiological sensible information with sub-second temporal accuracy and at the frequency higher than what has been commonly considered (i. e., ~ 0.1 Hz). Accelerated fMRI acquisitions hold the promise of detecting brain activity with high sensitivity and specificity.

Professor Fa-Hsuan the principal investigator if the Lab of Brain Imaging and Modeling, at National Taiwan University. He is currently visiting the Aalto University as FiDiPro Fellow.