Demonstrating network-level interactions between neural and vascular function

Our most recent work demonstrating that the brain’s vasculature is regulated to mimic neural networks is now published in Neuroimage: This work was performed in collaboration with Prof. Kevin Murphy’s lab at Cardiff University Brain Research Imaging Center in the UK.

Using fMRI, we simultaneously administer two neural paradigms (a working memory task and a visual stimulus) and one “vascular” paradigm that dilates blood vessels systemically (inhaled CO2). We then averaged together 30 fMRI datasets and used Independent Component Analysis to decompose the average dataset into network components. We readily identify three “neural networks” that show strong temporal correlation with the neural stimulus paradigms. These represent the Default Mode Network, Task Positive Network, and Visual Network – three robust and commonly observed functional brain networks expected to be activated or deactivated by our neural paradigms. However, we also see three additional components with similar network structure, and these three networks predominantly reflect the vascular stimulus design.

Our results demonstrate, for the first time, pairs of spatially similar neural and vascular brain networks. This suggests that the brain’s vasculature may be regulated to support specific brain networks, which must be taken into account to interpret fMRI studies of functional connectivity.

Ultra-high field MRI gives new insight into perfusion of lesions in Multiple Sclerosis

In our latest paper in European Radiology, we use ultra-high field MRI to explore the clinical usefulness of an emerging technique for measuring perfusion that doesn’t involve the injection of contrast agents. Instead, in Arterial Spin Labeling (ASL) scans, we magnetically label the blood and then image it as it flows into the tissue, allowing us to quantify perfusion. We developed this technique for a specific clinical application: to better understand the role of small cortical lesions in Multiple Sclerosis (MS).

In this technical development report, we present an optimized ultra-high-field ASL MRI acquisition that achieves high spatial resolution and low signal distortion. We assess the feasibility of using this imaging strategy to measure perfusion in MS cortical lesions, and demonstrate that our approach is sensitive to focal hypoperfusion in these small lesions. Because ASL MRI is safe and non-invasive, this type of imaging approach may facilitate the longitudinal study of acute lesion formation and development with frequent repeated scanning, allow us to test new therapeutic strategies, and give us better understanding of the heterogenous disease course in MS.

This excellent work was primarily accomplished by Rich Dury and Yasser Falah as part of their PhD research. For more details, the manuscript is available (open access) here:

New paper on multiparametric physiological imaging

A new article from the Bright Lab, titled “Multiparametric measurement of cerebral physiology using calibrated fMRI, has been accepted for publication in NeuroImage.

In the context of a calibrated fMRI experiment, we simultaneously characterize BOLD, blood flow, blood volume, tissue oxygen extraction, oxidative metabolism, and vascular reactivity throughout the brain.  Rather than focus on a single aspect of brain physiology, considering the relationships of multiple physiological parameters allows more accurate and precise characterization of neurovascular health in a range of clinical applications. This paper is part of a special issue on Physiological MRI, in collaboration with Paula Croal, Nic Blockley, and Dan Bulte at the University of Oxford.

View the full paper here.