A new analysis technique could get micron-scale data out of MRI scans of blood, brain cells, and even buried minerals–without having to re-engineer anything in the machines themselves. Standard MRI imagers today can resolve down to about half a millimeter on a live human, but in the 30 December print issue of PRL, a team describes a way of analyzing MRI data to squeeze out information on the average shapes and sizes of particles only a few microns across. Although the technique’s clinical feasibility remains uncertain, the researchers were able to accurately predict the MRI signal of red blood cells based on their size and shape.
MRI (magnetic resonance imaging) is based on the techniques of nuclear magnetic resonance (NMR). The scanner first aligns the nuclear spins of hydrogen atoms in the patient and starts them rotating in perfect concert. The nuclei emit maximum-strength electromagnetic waves at the start, but over time the rotating spins get out of synch, simply due to small differences in local magnetic fields. The unsynchronized spins cause the combined electromagnetic signal to decay with time, a phenomenon called relaxation.
MRI researchers can use the relaxation rates to identify different tissue types and even maladies, but the resolution is limited to about a millimeter. That limit comes from the system’s magnetic field gradient–the way in which the field ramps up from one end of the scanner to the other. The gradient allows the machine to label each point in space with a different value of the applied field. For higher resolution, the gradient would have to be steeper than is healthy for patients.
Valerij Kiselev, of Freiburg University Hospital in Germany, and Dmitry Novikov, a graduate student at the Massachusetts Institute of Technology in Cambridge, devised a new way of analyzing the relaxation rates in the case where small objects–cells or dust, perhaps–are suspended in solution. Local distortions in the magnetic field depend in part upon the shapes of the objects, and Kiselev and Novikov developed a theory that relates the relaxation decay with these shapes. Relaxation rates are complicated by the diffusion of the fluid around the objects, so they also incorporated a simple version of diffusion effects.
The team compared their theory with NMR data previously published by other groups, including studies of plastic beads in water and red blood cells in human blood. The theory successfully predicted relaxation rates based on the average shape and size of the objects, at least for the most dilute samples. According to the researchers, the same technique could be used to learn the micron-scale porosity of a mineral or whether micron-sized cells are shaped like discs, cylinders, or spheres–both at a hundred times the resolution of standard MRI machines.
Ronald Walsworth, of the Harvard-Smithsonian Center for Astrophysics in Cambridge, says the new method could potentially extend MRI in a new direction, but he cautions that this is only the first step in testing its practicality. “It shows that NMR definitely has this sensitivity to shape,” he says, “but can it tell a sickle-cell blood cell from a healthy one? The next step will be to start analyzing messier real-world systems.”