Detection of neuronal activity noninvasively and in vivo is a desideratum in medicine and in neuroscience. Unfortunately, the widely used method of functional magnetic resonance imaging (fMRI) only indirectly assesses neuronal activity via its hemodynamic response; limiting its temporal and spatial accuracy. Recently, several new fMRI methods have been proposed to measure neuronal activity claiming to be more direct and accurate. However, these approaches have proved difficult to reproduce and are not widely applied mainly because of a dearth of “ground truth” experiments that convincingly establish the correlation between the magnetic resonance (MR) signals and the underlying neuronal activity. In addition, limited knowledge of water dynamics in living tissue restricts our understanding of the underlying biophysical sources of these candidate fMRI signals.

To address the first problem, we developed a novel test system to assess and validate fMRI methods, in which real-time fluorescent intracellular calcium images and MR recording were simultaneously acquired on organotypic rat-cortex cultures without hemodynamic confounds. This experimental design enables direct correlation of the candidate functional MR signals with optical indicia of the underlying neuronal activity. Within this test bed, MR signals with contrasts from water relaxation times, diffusion, and proton density were tested. Diffusion MR was the only one shown to be sensitive to the pathological condition of hyperexcitability, e.g., such as those seen in epilepsy. However, these MR signals do not appear to be sensitive or specific enough to detect and follow normal neuronal activity.

Efforts were made toward improving our understanding of the water dynamics in living tissue. First, water diffusivities and relaxation times in a biomimetic model were measured and quantitatively studied using different biophysical-based mathematical models. Second, we developed and applied a rapid 2D diffusion/relaxation spectral MR method, to better characterize the heterogeneous nature of tissue water. While the present study is still far from providing a complete picture of water dynamics in living tissues, it provides novel tools for advancing our understanding of the possibilities and limits of detecting neuronal activity via MR in the future, as well as providing a reproducible and reliable way to assess and validate fMRI methods.