is a freelance medical writer in Mamaroneck, NY.
Full Text (PDF)
Current and developing modalities of medical imaging allow clinicians and researchers to obtain clearer pictures of the inner workings of the human body with goals that include improving diagnosis and treatment, developing new therapeutics, and understanding how the body and mind work in health and disease.
Heart of the MatterAccording to Robert C. Hendel, Nuclear Cardiologist, Midwest Heart Specialists, Fox River Grove, IL, cardiac computed tomography (CT), a non-invasive technology for cardiac imaging “is the most explosive growth technology we have. With 64-slice technology, we now have the ability to get beautiful, wonderful pictures of the coronary arteries.” He notes that the technology is ahead of controlled studies to determine how best to use it, and in what circumstances. Guidelines were expected this summer from the America College of Cardiology on the appropriateness of the use of CT angiography.
At the moment, there is agreement that CT angiography should not be used for routine screening, but rather to evaluate symptoms in candidates for further treatment and in patients with risk factors (e.g., someone with unanswered questions after undergoing a stress test). “I am pro-CT. I think it is great,” Hendel says, “but we need to be circumspect in how to apply new and expensive diagnostics. CT angiography is exciting and has its place. The equipment is expensive and is not for every hospital, certainly not for private practice, and will never be used bedside.”
Image 1.
All in the Mind
Randy L. Buckner, Professor of Psychology, Harvard University, Cambridge, MA, and a Howard Hughes Medical Institute Investigator, is using functional magnetic resonance imaging (fMRI) of brain activity to study human memory, “an incredibly powerful tool, which allows one to see the brain in action.” Cognitive activity can be related to blood flow in the brain. When normal adults lie in a scanner musing to themselves, there is a lot of structured activity in the brain that he refers to as the default network or mode. The most active regions are similar in healthy people. In patients with Alzheimer's disease, one of Buckner's research interests, the areas of accumulation of amyloid plaques, associated with the disease pathology, show a pattern similar to the default pattern. “I'm excited about going forward,” he says, “and seeing the emergence of more specific and diverse aspects of brain function, structure, activity, and molecular events.” He expects the combination of several different brain imaging methods in the future may allow a more complete picture of brain function, as each method has its limits. This may lead to better clinical diagnoses and earlier disease detection.
Jessica Turner, Specialist, Department of Psychiatry and Human Behavior, University of California (UC), Irvine, is also the Project Manager of Function BIRN. Biomedical Informatics Research Network (BIRN) comprises three test bed projects, Function BIRN, Morphometry BIRN, and Mouse BIRN. BIRN is sponsored by the National Institutes of Health's (NIH) National Center for Research Resources (NCRR) and promotes advances in biomedical and healthcare research through the development and support of a cyberinfrastructure that facilitates data sharing and multi-institutional collaboration. The mission of Function BIRN is to understand the underlying causes of and develop new treatments for schizophrenia. UC Irvine is one of 14 laboratories participating in Function BIRN. Turner points out that “BIRN was a technology looking for an application. We were looking ahead toward ideas. Could we pool our data and find out more?” Some of the questions being asked include what are the differences when neuroimaging results from different laboratories disagree? How are data best collected? How can individual researchers understand what another researcher did? “How to share and combine data,” Turner observes, “is not a scientifically interesting question, but it is a scientifically necessary question.” Turner believes Function BIRN will help further individualized medicine and allow better diagnosis, treatment, and drug development. Better brain imaging and the ability to compare data across laboratories will provide information on how treatment works and ultimately why it works in one person and not in another. This will require asking new questions and looking at large numbers of people to acquire the necessarily large data set. One exciting first that Function BIRN has accomplished is to scan the brains of a group of healthy subjects in different scanners at different locations around the U.S. and compare the results. These data will allow the development of methods to correct for data collected using different scanners at different locations.
Image 2.
Putting It Together
Laurie Loevner, Professor of Radiology at the Hospital of the University of Pennsylvania, Philadelphia, PA, is using new imaging technology that combines CT with positron emission tomography (PET) for earlier diagnosis of cancer and detection of metastases. “Hot spots” of radioactive isotope uptake seen by PET scan, which identify areas of metabolic activity associated with malignancy, are localized anatomically using CT imaging. “Not all hot spots are cancer,” notes Loevner, and this technology allows “segregation of normal and abnormal.” Currently, PET/CAT relies on computer-generated analysis to reconstruct images. Advantages of PET/CAT include improved localization of lesions in obese patients and localization of additional lesions that cannot be detected using conventional image reconstruction. Loevner predicts that PET/CAT will be used to assess response to cancer treatment and to target radiation therapy. Currently, most reimbursement for PET/CAT is for staging and diagnosis, not treatment. “We can use this technology to determine treatment options and improve patient quality of life. Longevity is important,” she says, “but we must also think about the quality of existence of those living with disease.”
Image 3.
On the Way to the Clinic
Other imaging modalities are in development. For example, Marianne Manchester, Associate Professor, Department of Cell Biology, Scripps Research Institute, La Jolla, CA, and colleagues are using viral nanoparticles as the basis of noninvasive in vivo vascular imaging probes. “Viruses are coming to the forefront because of nanoparticle toxicity,” Manchester says. “When we started this work in 2002, we were far behind the other technologies.” Now, however, viral nanoparticles seem to have an advantage over conventional nanoparticles in their toxicity, bioavailability, and pharmacokinetic profiles. “It's a challenging area,” she observes, “but the rewards are immense, so it's worth the effort.”
QuantomiX™ is developing lipid imaging technology. Lipids have traditionally been assessed by solvent extraction, which does not allow tissue- and organelle-specific localization. Their technology allows lipid imaging using the electron microscope (EM) on wet, encapsulated samples rather than dry samples exposed to a vacuum. Ory Zik, QuantomiX CEO, predicts that, within several years, lipid imaging in cells and tissue samples will be used for early diagnosis of inflammation by measuring lipids in white blood cells in the intensive care setting and not long after for point-of- care diagnosis of insulin resistance by measuring early accumulation of lipids in skeletal muscle. “We're making EM, the least practical but most accurate imaging tool, more practical,” Zik says. He predicts that miniaturization will eventually result in an affordable, desktop EM requiring minimal sample preparation.
Future DirectionsBruce R. Rosen, Director of the Center for Functional Neuroimaging Technologies at the Athinoula A. Martinos Center for Biomedical Imaging at Massachusetts General Hospital, Charlestown, MA, sees some broad trends in the field. One is the combination of modalities, such as PET with CT, MRI with PET, and soon CT with MRI and PET. “No single examination has all the answers,” he observes. “The development of single devices allowing several types of scans to be performed in one sitting should solve the problem of registration. Instead of having to put images together, the scanner will use the data from two or more modalities to create a common reconstruction.” It is an interesting and challenging research and development puzzle, he concludes. A second general question that will have a bigger impact in the future is how to meet the need for computational tools to extract and quantify data in a timely and cost-efficient manner. “What does a radiologist do with a data set of 20,000 images in the few minutes before the next patient?” he asks. An effective computational tool would extract the critical features from these images and alert the radiologist to these key features and how they differ from known controls. The ability to extract and compare data for a patient from visit to visit or to known controls will require a lot of work, but in the end, he says, it will lead not only to better and more sensitive information, but will also guide the development of new treatments.
Image 4.
Turner believes that in a few years, she will be able to sit down at a web site and be able to find all the fMRI data on left-handed schizophrenics who did a working memory task. For the moment, useable tools have to be built from the ground up, but Turner is confident that within 5 years, the databases that will allow such searches will be commonplace.
