Tractometry. (360G-Wellcome-096646_Z_11_Z)

Overview Nothing defines the function of a neuron better than its connections, so wrote Mesulam in his preface to Fiber Pathways of the Brain (Schmahmann and Pandya, Oxford 2006), highlighting the importance of the white matter to the functioning of the brain. Following Mesulam, I will show that: Nothing defines the ability of two cortical regions to communicate better than the microstructure of their connections. My vision is three-fold: (1) I will introduce, develop and disseminate tr actometry16 and it will become commonplace in all neuroimaging studies, so white matter microstructural assessment will be promoted to equal footing with grey matter functional imaging; (2) as a consequence, researchers will abandon DT-MRI and its blunt index of fractional anisotropy in favour of far more compartment-specific white matter indices (myelination, axon morphology); (3) tractometry will be instrumental in advancing our understanding of brain function in health and disease. Background: Despite a long appreciation that White Matter Matters(17), and has a fundamental role in behaviour, cognition and perception, the development of EEG, PET and fMRI led to research focus on grey matter, the white matter being effectively treated as a homogenous substance in imaging studies. In 1994, my former NIH-mentor, Peter Basser, described DT-MRI which first allowed access to tissue microstructural information in vivo. The three-dimensional displacement profile of water can b e hindered by cell membranes and cellular inclusions and DT-MRI models this as a unimodal Gaussian (tensor), yielding two widely used indices: (i) the principal eigenvector, which is used as a proxy estimate for fibre orientation to reconstruct white matter pathways (tractography); and (ii) fractional anisotropy (FA) which is used as a quantitative index of tissue microstructural organization. Mathematically defined as the standard deviation of the eigenvalues divided by the root-mean-square, F A is zero when the rate of diffusion is the same in all directions and one, when diffusion is constrained to move along a single axis. FA-mapping and tractography have since been used extensively (a PubMed search on fractional anisotropy returns >2000 papers in diverse applications, and http://www.ncbi.nlm.nih.gov/pubmed?term=tractography yields >1000 papers). Frequently, changes in FA are interpreted (erroneously) as changes in integrity and differences in probabilistic tracking as change s in connectivity. Several studies have correlated performance on a cognitive test with FA in specific white matter tracts(24). Intriguingly, there is evidence that experience over years(18)/months(19) and, importantly, that brain training(20) can modify white matter FA, which has been interpreted as increased myelination. Finally, with a connectionist approach to understanding the brain, there is increasing interest in integrating white matter information into models of functional/effective c onnectivity. However, with few notable exceptions(21), very little progress has been made in this regard, THE PROBLEM I have discussed the fundamental problems with FA-mapping/ tractography for understanding brain function (22) (cubric.psych.cf.ac.uk/people/derek-jones/Jones%20-%20Imaging%