Invention and Development of Diffusion Tensor MRI (DT-MRI or DTI) at the NIH

At the time of the invention of DT-MRI (or DTI), I was employed as a Staff Fellow (a postdoctoral fellow position) in the Mechanical Engineering Section (MES) of the Biomedical Engineering and Instrumentation Branch (BEIB) in the Intramural Research Program (IRP) of the National Center for Research Resources (NCRR) of the National Institutes of Health (NIH). (People in the federal government like acronyms!) This was my fi rst postdoctoral position after having graduated from Harvard University’s Division of Applied Sciences, where I had studied biofl uid and continuum mechanics, and applied mathematics. Initially, my research activities within BEIB focused on “biomimetics,” specifi cally, the development of models and model systems that mimic more complex biological ones; and studying the response of tissues to impressed mechanical and electric fi elds. By the early 1990s, my two primary projects involved understanding the physical basis of electromagnetic stimulation (i.e., “magnetic stimulation”) of the peripheral (PNS) and central nervous systems (CNS), and understanding the mechanical effects (swelling and consolidation) in the brain during drug infusion. Bradley Roth (a talented biophysicist and fellow Staff Fellow in the BEIB) and I had developed a mathematical model of magnetic stimulation to explain how induced electromagnetic fi elds in neural tissue can cause depolarizing or hyperpolarizing transmembrane currents in long axons in the PNS. In one of our papers (Basser and Roth, 1991 ), we described an axon’s trajectory or path as a 3D space curve — a concept that later resurfaced in DTI tractography (Basser, 1998 ) — while considering the case of magnetic stimulation of an axon whose path is kinked or sinuous. Developing these models of magnetic stimulation led me to realize that we would never be able to determine the action of induced electromagnetic fi elds on nerves in the CNS or the PNS if we 1) didn’t know the coordinates and trajectories of the nerve pathways of interest, and 2) more generally, didn’t know the anisotropic and heterogeneous electrical properties (i.e., the Summay box