Shedding Light on Brain Activity and other exciting adventures in preclinical Magnetic Resonance Imaging

High field, 7 Tesla, preclinical Magnetic Resonance Imaging finds application across the biosciences. It can be used to distinguish cancerous tissue from normal tissue, and validate disease phenotypes in transgenic and xenograft small animal models. It can be used to assess efficacy of selective therapeutics that are able to target diseased organs (brain, heart, liver, kidney, muscle, and bone etc.) in animal models. Furthermore, the technique can map brain function and as such functional magnetic resonance imaging (fMRI) has become the cornerstone of cognitive neuroscience in recent years.

The widely used Blood Oxygenation Level Dependent (BOLD) signal is often used to interpret changes in neuronal activation. However, at present, a biophysical understanding of the neurovascular drivers of the BOLD signal is not clear and thus hinders any quantitative estimation of the underlying neuronal activity. My neuroimaging research drives towards building a forward biophysical model of this complex relationship.

Validation and refinement of the haemodynamic response models underlying fMRI signals, an essential precondition for the correct interpretation of human BOLD data, requires invasive multimodal preclinical imaging. I have developed an innovative in-vivo methodology for concurrent fMRI and 2D optical imaging spectroscopy (2D-OIS) techniques for simultaneous measurement of BOLD signal and underlying haemoglobin changes to neuronal activation in the healthy rodent model (Kennerley et.al. 2012). Current research centres on the mathematical development of optical imaging, specifically light transport through tissue models based on Monte Carlo methods, parametrised by 3D MRI data, to extract layered brain function information with light alone.  

Applications of this technology include: 1) understanding of the negative BOLD signal (Boorman et al 2010); 2) refinement of mathematical and biophysical models of both the BOLD signal and optical imaging spectroscopy techniques (Kennerley et.al. 2009); 3) calibration of non-BOLD fMRI techniques such as VASO and ASL; 4) interpretation of abnormal BOLD responses; specifically in disease and trauma conditions in which either neuronal or haemodynamic breakdown could be responsible. Only in furthering the development of forward biophysical models of brain activity will be it possible for fMRI techniques to truly read the mind.

[1] Kennerley et.al. (2012) NeuroImage 61(1): 10-20;

[2] Boorman, L.W. et.al. (2010) Journal of Neuroscience. 30(12): 4285-94;

[3] Kennerley, A.J. et.al (2009) NeuroImage 47:1608-1619;

More on Dr Aneurin J Kennerley