Researchers Present New Physics-Based Medical Imaging Techniques

Medicine has long drawn on physics-based techniques for a variety of imaging, measurement and testing applications. Several researchers described the latest developments in this area at the 2002 APS March Meeting in Indianapolis.

Micro-tesla MRI was reported by Robert McDermott, a member of John Clarke's group at UC Berkeley. The principle behind MRI is nuclear magnetic resonance (NMR), a process in which a magnetic field is used to orient atomic nuclei in space while a burst of radio waves explores the nuclear energy levels by charting the frequencies at which energy is absorbed resonantly.

In addition to establishing chemical identity, NMR can also be turned into an imaging method by carefully watching the timing and the location of the re- emitted radio waves.

A tumor, say, will have a slightly different water density from surrounding healthy tissue. Computer processing and contrast enhancement will disclose the tumor's position to a trained observer. Generally large magnets are required to produce sharp NMR images, and the development of a low-field version would benefit medical and scientific studies.

Mark Haacke of the MRI Institute for Biomedical Imaging in St. Louis discussed a new MRI technique called susceptibility weighted imaging (SWI). The technique measures differences among brain tissue in its magnetic susceptibility, essentially its magnetic response to the applied magnetic field of the MRI machine. Yielding unique information from veins and blood products, SWI has already provided more sharply detailed MRI images of blood vessels in the brain than previously possible and the presence of small hemorrhages in heretofore unavailable detail.

SWI can potentially detect angiogenesis, the growth of blood vessels caused by cancer, and may improve diagnosis of Parkinson's and Alzheimer's diseases, through its ability to monitor iron deposits in the brain.

Electrical measurements of individual living cells can provide powerful information without the use of optical techniques, which often require labeling the molecules of interest with fluorescent markers. Cells are not the clean, compartmentalized units depicted in high-school biology textbooks, but rather complex networks of interacting molecules which require new tools to be studied in living detail.

Proteins in particular are of great interest because they are the molecules that perform the most reactions in the cell, from metabolism to DNA replication. But while whole-genome technologies such as DNA microarrays can monitor quantitatively the relative abundances of all the mRNA species within a cell, they cannot take inventory of a cell's content's in vivo.

Towards these ends, Lydia Sohn of Princeton University described several electrical biosensors at the APS meeting, including one that can measure the amount of DNA in single living cells. Passing through a small fluid chamber in between two metal electrodes, a cell changes the electrodes' capacitance in a way that is directly proportional to the amount of the cell's DNA (which carries a negative charge). This technique can potentially identify the stage of a cell's development and also distinguish cancerous cells from normal ones.

Omar Saleh, part of Sohn's group at Princeton, also presented an artificial pore, a microchip device that can determine the size of a tiny object (such as a cell) by detecting changes in electric current as the object passes through a tiny opening, or pore in a fluid chamber containing a pair of electrodes.

The ultimate goal of Sohn's lab is to be able to take inventory of a living cell's protein contents, something that cannot be done with current protein assay techniques, which require the lysing (destruction) of cells. Ideally, they would like to "watch" the proteins "interplay" with one another in the so-called protein network.

Despite its name, the AFM (Atomic Force Microscope) does not produce atomic- resolution images of proteins or other large molecules. To extract more detailed information from AFM images of macromolecules, one can directly subtract the effects of the tip but the results are often inaccurate.

Steven Eppell and Brian Todd of Case Western Reserve University presented a new technique for obtaining submolecular information about proteins. Investigating aggrecan, a cartilage protein important in osteoarthritis, the researchers used a technique that combined AFM with genome information and transmission electron microscopy data.

All of the data were integrated by using a sophisticated image processing technique to provide a best guess at the 3D structure. The resulting refined structure yielded new information on the molecule, showing distinct locations of kinks as well as regions of mechanical flexibility.

The researchers hope to combine their results with AFM-measured force fields around cartilage proteins to link the biological and mechanical properties of cartilage with its molecular structure. This approach has the potential to provide information on molecular-scale mechanisms for arthritis and lead to intelligent drug design and other interventions to prevent or alleviate the disease.