UCSF’s neuroradiologist Alisa Gean, M.D., writes “DTI ready for prime time?” in response to recent publicity of radiology’s newest rage – high-definition fiber tracking MRI for TBI.
Although new probabilistic methods for DTI evaluation of nerve fiber bundles are providing unique information about the microstructure of white matter, the best approach to maximize information from DTI is still a matter of debate, especially when it comes to quantifying the integrity of nerve fiber bundles. Visual inspection
alone of the images can be unreliable. Moreover, while DTI is providing some remarkable insights into TBI, some believe that there is an “irrational exuberance” for it in the literature, especially for 3D color tractography (). Abnormalities within the tractogram, while visually very appealing, are not specific for TBI. DTI findings must be interpreted in the context of the characteristics of the injury, clinical features, other associated imaging findings, and co-morbid
conditions. Additionally, the white matter tracts generated with DTI
tractography are quite variable and depend on the parameters used to create the tracts. Indeed, the sensitivity of fiber-tracking algorithms to many physical and computational variables is still poorly understood, and their behavior in the face of injured tissue even less so. Numerous issues must still be accounted for to reduce false positives and false negatives when interpreting DTI studies. For example, FA and ADC are affected by imaging variables such as field strength and resolution, as well as subject variables including pre-existing disease. Patient age affects the interpretation of FA,
as low FA may be expected during early development and enescence. Another limitation of DTI is that it samples only a very small subset of the full diffusion information encoded in q-space and describes
diffusion as single compartment (i.e., Gaussian). This assumption, however, falls short in fiber crossings and in biological tissue, which may exhibit restricted, non-Gaussian diffusion (). Diffusion Kurtosis Imaging (DKI) is a new extension of conventional diffusion imaging that measure water diffusion in non-Gaussian tissue so it is thus more accurate than DTI. Kurtosis is a dimensionless statistical measure of the departure of a distribution from Gaussianity. DKI is thought to be an imaging marker of tissue structure complexity (i.e., cellular compartments and barriers). DTI is also limited by the fact that it suffers from severe artifacts in the presence of magnetic field inhomogeneities. In addition, the significant T2* decay that can
occur during long echo trains makes high-resolution acquisitions challenging. With the recent advance of parallel imaging technology, less spatial distortion and higher signal-to-noise ratio can be achieved in a time-efficient manner with single-shot EPI. More case-control studies are necessary, however, with age-matched comparisons of DTI results between patients and corresponding
control subgroups. Finally, the difficulties of post-processing and the need for statistical analyses to evaluate the effects on FA maps of spatial distortion correction induced by gradient nonlinearity, misregistration, image processing, and variable protocol parameters (e.g., b value, number of diffusion-encoding gradient directions, number of excitations, software data analysis package, etc.) are currently keeping DTI primarily in the research arena. Therefore,
further research with comprehensive analysis needs to be done before there is widespread adoption of this new technique into our routine clinical imaging armamentarium. DTI and DKI are
further addressed in Chapters 4 and 5.
 Field AS. Diffusion tensor imaging at the crossroads: fiber tracking meets tissue characterization in brain tumors. AJNR Am J Neuroradiol. 2005 Oct;26(9):2183-6.
 Menzel MI, Tan ET, Khare K, et al. Accelerated Diffusion Spectrum Imaging in the Human Brain Using Compressed Sensing Magnetic Resonance in Medicine 000:000–000 (2011)