Caltech Brain Imaging Center Seminar

It's been about 32 years since the first functional MRI (fMRI) signals were observed and since then, as it has grown in sophistication and utility, the number of groups using it as their primary investigative tool has grown. Likewise, the range and depth of neuroscience questions addressed has also increased. Advances have occurred in acquisition approaches, processing methods, our insight into the mechanisms behind the signal, and applications. In this talk, this progression will be traced, highlighting some of the more significant advances. The bulk of this talk will be devoted to the question of where fMRI is going. What are the major challenges and most promising opportunities or direction in the field today? How will the method achieve more clinical traction?

Specifically, the future of several areas of fMRI will be highlighted. Among these are: the growth of large, shared databases, ultra-high resolution fMRI - specifically cortical depth-resolved fMRI, the development of novel pulse sequences like multi-echo fMRI and simultaneous contrasts, the potential of deep sampling of data to draw more subtle inferences and to delineate more clearly differences and similarities between individuals, the use of voxel-wise predictive modeling as well as intersubject correlation approaches, real-time neurofeedback, multi-modal integration, novel and perhaps direct neuronal contrasts, and the most compelling immediate clinical applications.

Also discussed are challenges that have been present since the beginning such as image distortion, physiologic noise and motion, the confining space and high acoustic noise associated with MRI, and the ubiquity of variable hemodynamics as we continuously move towards higher spatial and temporal resolution. Lastly, I will conclude with my own projections of where fMRI might be in the next 5, 10, and 20 years in terms of dissemination, technology, and utility.

This seminar is part of the "Future Advances in Multi-modal Neuroimaging Methods" series. Existing methods for measuring human brain function non-invasively such as fMRI, MEG and EEG have helped to considerably advance our understanding of human brain function over the past several decades. However, these methods suffer from substantial limitations in spatial and temporal resolution, thereby imposing major constraints on the kinds of neuroscience research questions that can be addressed in humans. This seminar series explores cutting-edge new technologies and methodologies for multi-modal imaging. We will ask whether such new approaches have the potential to overcome existing limitations and open up new research directions in human neuroscience over the coming decades.