Fig. 1.
Schematics of current standard model of brain microstructure (a) and the novel model proposed in this work (b). Current conjecture envisions the tissue component in an MRI voxel as subdivided into two non-exchanging compartments: intra-neurite and extra-neurite space. The total MRI signal is then given by the weighted sum of the signals from water molecules diffusing in each compartment, with relative signal fractions fin and 1-fin, respectively (a). We propose a new picture: the tissue component of an MRI voxel is subdivided into intra-cellular and extra-cellular non-exchanging compartments. The total signal is the weighted sum of the signal from water molecules diffusing in each compartment, with relative signal fractions 1-fec and fec, respectively. Furthermore, the intra-cellular compartment is itself divided into two non-exchanging sub-compartments: intra-neurite and intra-soma. The intra-cellular MRI signal is then given by the weighted sum of the MRI signal from water molecules diffusing within the two sub-compartments, with relative signal fractions fin and 1-fin, respectively (b).