HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Thrall, J. H. Search for Related Content PubMed PubMed Citation Articles by Thrall, J. H.

Building Research Programs in Diagnostic Radiology Part II. Basic Research¹

¹ From the Department of Radiology, Massachusetts General Hospital, MZ-FND 216, Box 9657, 14 Fruit St, Boston, MA 02114. Received October 11, 2006; final version accepted October 17. Address correspondence to the author (e-mail: thrall.james{at}mgh.harvard.edu).

Basic research in radiology can be broadly subdivided into two categories: Research directed at furthering the state of the art of imaging—new imaging methods, new pharmaceuticals, and new image-guided interventions—and research that employs imaging methods as tools for biomedical and bioscientific discovery. Both directions offer tremendous opportunities and are worthy of support. While there are substantial points of convergence, the nature of research in these two subdivisions is quite different, having fundamentally different purposes and attracting investigators with different backgrounds and interests. The strategies, resources, and approaches employed for research program building must be tailored accordingly.

	BASIC RESEARCH AIMED AT FURTHERING THE STATE OF THE ART OF MEDICAL IMAGING

TOP INTRODUCTION BASIC RESEARCH AIMED AT... AN ACADEMIC CONUNDRUM IMAGING METHODS AS TOOLS... CHARTING A PATH FORWARD CONCLUSION References

 
Research aimed at furthering the state of the art of imaging is founded in physics, engineering, materials science, computer science, and mathematics. These are the basic scientific disciplines that underwrite the basis for discovering and developing new imaging methods and for inventing new or improved imaging devices. The development of diagnostic imaging pharmaceuticals adds the disciplines of chemistry, molecular and cell biology, genetics and/or genomics, and pharmacology, among others. Research directed at inventing new image-guided therapies may encompass any subset of these disciplines.

The most fundamental observation about medical imaging is that every imaging method entails the recording of energy emanating from biologic tissues, internally or externally, and, in most applications, the exposure of tissues to energy. Thus, the most fundamental or basic step in inventing a new imaging method is the discovery of a form of energy suitable for creating medical images and exploration of the nature of the interactions of that energy form with biologic tissues—radiobiology in its broadest sense. The discovery of a suitable energy source for medical imaging was precisely the contribution of Conrad Roentgen more than 100 years ago; that source is one we are still exploring and exploiting today.

New energy sources continue to be added to the imaging armamentarium. Understanding their characteristics and optimal integration into practical imaging methods is an enormous ongoing opportunity in basic research. For example, a number of "reporter" molecules with fluorescent emissions in the near-infrared spectrum have been adopted for use in molecular imaging (1,2). Why choose agents fluorescing in a small portion of the near-infrared spectrum (wave length, approximately 700–900 nm) instead of in the visible light spectrum (wave length, approximately 400–700 nm) as is commonly done for in vitro molecular imaging of histologic slices (eg, green fluorescent protein has an emission wavelength of 508 nm)? Very simply, the absorption coefficients for hemoglobin and deoxyhemoglobin in tissues are unfavorable in the visible light spectrum and are much less dominant in the near-infrared domain (1,2). Choosing the near-infrared spectrum versus the visible light spectrum reduces photon absorption in tissues and increases photon flux and, therewith, useful tissue depth for imaging.

With an energy source at hand, basic research in radiology is then directed at inventing the means to apply the energy as well as to record, store, transmit, display, manipulate, and analyze the patterns in the energy distribution emanating from the exposed tissues. Each of these steps in the imaging sequence is a rich subject for basic research aimed at inventing or optimizing imaging methods (3,4). After all, computed tomography (CT) is fundamentally just another way to administer and record x-rays, an energy source that had been in use for numerous years before the CT technique was invented.

An important component of basic research in imaging is studying the safety of the respective methods for the purpose of understanding risks and for undertaking design improvements to reduce risk. The common mechanism of injury from imaging is the deposition of energy in tissues at a rate or aggregate amount too great for its negative effects to be dissipated. It is interesting to note that many of the same energy sources are used for both diagnostic and therapeutic purposes, including image-guided therapy. The strategy in therapy, more or less, is to administer as much energy as possible without causing undue collateral damage to adjacent normal tissues. The strategy for diagnostic applications is to administer as little energy as possible for accurate diagnostic inference—the ALARA (as low as reasonably achievable) principle. Given the amount of publicity and controversy currently being generated about the amount of radiation received by patients undergoing CT scanning (5), the imaging community must increase its efforts (6) to understand how to optimize the balance between medical value created and medical risk induced.

The invention of imaging agents—radiopharmaceuticals, optical beacons, and x-ray and magnetic resonance (MR) imaging contrast agents—is another pillar of basic research activity in radiology. Fundamentally, medical value is derived in imaging by understanding the linkage between the patterns of energy distribution emanating from tissues and the state of the exposed tissues—healthy or diseased. Diagnostic imaging agents increase the information content of medical images by helping to determine the spatial and temporal distribution of energy reaching an imaging-detector system in ways that enhance diagnostic inference. For example, iodinated x-ray contrast media alter the energy distribution in the recorded image by absorbing a portion of the energy in the x-ray beam. These alterations permit diagnostic inference either directly, as in the event of a visualized stenosis or blockage in a blood vessel, or indirectly, through kinetic patterns of altered tissue density in tumors. Radiopharmaceuticals determine the recorded patterns of energy in nuclear medicine images through their respective biodistributions in tissues, which can be linked to the presence or absence of disease.

Diagnostic pharmaceuticals have design requirements and, hence, research requirements not inherent to other pharmaceuticals. Classical pharmacology uses the terms pharmacokinetics to describe the time course of pharmaceutical distribution and pharmacodynamics to describe the desired therapeutic effect of a drug, but does not have a term that applies to the unique purposes of diagnostic imaging pharmaceuticals in altering the temporal and spatial distribution of energy emanating from tissues. Perhaps the term pharmacophysics could be adopted to reference the study of these energy-related effects that are unique to diagnostic imaging agents. The example of selecting near-infrared versus visible light agents for in vivo molecular imaging illustrates the concept. Pharmacophysics is the special domain of imaging scientists and another unique opportunity for basic imaging research.

The biology controlling the distribution of imaging agents is a subject area where imaging research intersects broadly with many other domains in biomedical research. Each imaging agent may be thought of as having two components—one conferring the desired energy effects (pharmacophysics) for detectability, and one conferring tissue localization. The study of how to achieve desired tissue localization encompasses the principles of pharmacology, again optimized for a given imaging application. Vascular dynamics, the functioning of receptor and enzyme systems, antigen-antibody interactions, and the expression of genes help inform the design of the localization properties of imaging agents. The development of imaging agents benefits from the enormous amount of work on these subjects taking place outside radiology and helping to define targets and mechanisms of distribution and localization for imaging agents. Rather than trying to duplicate this work, this knowledge can often be imported into imaging research programs for use in crafting diagnostic pharmaceuticals.

	AN ACADEMIC CONUNDRUM

TOP INTRODUCTION BASIC RESEARCH AIMED AT... AN ACADEMIC CONUNDRUM IMAGING METHODS AS TOOLS... CHARTING A PATH FORWARD CONCLUSION References

 
The uniqueness of so much of the basic research done in radiology compared with most biomedical science brings with it a number of challenges. First, most biomedical researchers outside radiology have no knowledge of the disciplines underlying much of the basic research in radiology. These scientists are principally biologists—not engineers, mathematicians, and physicists. They do not know the lexicon of the imaging sciences and generally have no way to evaluate the quality or impact of research aimed at building the imaging toolkit. For example, CT is a tour de force in mathematics, physics, and engineering, not a shred of which is of interest to biomedical academicians outside the specialty of radiology. Yet it is precisely research founded on mathematics, physics, and engineering that results in new imaging methods and that forms the basis of so much progress in our field.

At our medical schools, this uniqueness of basic imaging research has created challenges in securing promotions for imaging scientists and radiologists. Conversations with radiology chairmen around the country indicate that others have faced the same difficulty in the promotions process. At Harvard Medical School, the radiology chairmen have met with the dean to attempt to explain that the focus of a career in imaging research might be on the exploration of new energy sources, image reconstruction algorithms, or a particular imaging method, rather than on a disease or a biologic system, both of which are the dominant themes of most careers in biomedical research. It would benefit the academic radiology community to find a way to document the prevalence of this conundrum and perhaps prepare a white paper that could be used to help inform promotions committees in our medical schools about the nature of scholarship in basic research in radiology.

A different wrinkle in this conundrum has faced basic imaging scientists in interactions with the National Institutes of Health (NIH) over the years. At the NIH, a focus on organ systems, diseases, and conditions is the dominant organizational model defining the interests of the respective institutes. The institute structure does not match well with the disciplines and activities involved in basic research into new imaging methods. This has historically placed imaging scientists at a substantial disadvantage in securing grants. This historical challenge was partly addressed with the establishment of the imaging program in the National Cancer Institute and more fully addressed in 2000 with the establishment (7) of the National Institute for Biomedical Imaging and Bioengineering (NIBIB). The latter was established largely through the efforts of leaders in radiology and through the work of the Academy for Radiology Research. The NIBIB is now a primary home at NIH for basic research directed at the development of new imaging methods.

	IMAGING METHODS AS TOOLS FOR BASIC BIOMEDICAL RESEARCH

TOP INTRODUCTION BASIC RESEARCH AIMED AT... AN ACADEMIC CONUNDRUM IMAGING METHODS AS TOOLS... CHARTING A PATH FORWARD CONCLUSION References

 
As rich as the landscape is for the development of new imaging methods and new imaging pharmaceuticals, an even larger opportunity presents itself through the use of imaging methods as tools or probes for the study of questions of far-ranging and general biomedical interest. The watershed for this was reached around 1990 through the convergence of better and more powerful analytic methods, the invention of functional MR imaging (8), the astonishing expansion of molecular imaging far beyond its origins in nuclear medicine (1), and the transition in radiology from analog to digital systems (9) for image capture and management. Before then, with the notable exception of positron emission tomography (PET), the methods offered by medical imaging were of very limited value in basic biomedical research and therefore of little interest to scientists in disciplines outside radiology. The contemporary trend of wide-scale adoption of imaging methods by basic researchers outside radiology constitutes a kind of translational activity that might be characterized as from "bench to bench" rather than from "bench to bedside."

Examples are plentiful. There is no question that functional MR imaging has revolutionized basic research in cognitive neuroscience, with major effects on basic research in neurology, neurosurgery, psychiatry, and psychology. Brain mapping has been performed for years in experimental animals but, beyond limited clinical applications in surgery, was not feasible in humans until the availability of functional MR imaging. More importantly, the human mind is the basis of human identity and, before functional MR imaging, was only accessible for study through subjective means. Scientists can now literally watch the human mind at work processing thoughts and emotions. For the first time, scientists can objectively and reproducibly study pain-and-reward circuitry (10). Functional MR imaging is now also being used in genomics research to create imaging phenotypes (11) that facilitate discovery of genetic polymorphisms. It is almost unimaginable today for any institution with major interests in the neurosciences or psychology to not have substantial functional MR imaging capability.

Molecular imaging has far-reaching ramifications as a tool in scientific discovery because of its extraordinary versatility. Literally any tissue can be targeted. Optical imaging is proving especially valuable in small-animal models of disease, where tissue depth is less challenging than in humans, and this method is being adopted by investigators across the basic biomedical research community. Optical imaging agents are relatively easy to make and lend themselves to amplification strategies yielding high target conspicuity (1,2). Targeting of overexpressed enzymes is a particularly useful approach, because agents can be readily tailored to the expression profile of the particular disease or condition being investigated.

The ability to follow disease progression sequentially over time in expensive animal models is a great advantage of imaging and often yields information not obtainable by sacrificing an animal. Every institution involved in biomedical research needs to address how to support molecular imaging applications and small-animal imaging facilities more generally.

Molecular and functional imaging methods served as the triggers for the paradigm shift toward the use of imaging methods as tools in basic biomedical research and have the highest profiles. Many other newer imaging methods are proving useful as well, including high-resolution CT, intravascular ultrasonography, diffuse optical tomography, and MR spectroscopy, among others.

	CHARTING A PATH FORWARD

TOP INTRODUCTION BASIC RESEARCH AIMED AT... AN ACADEMIC CONUNDRUM IMAGING METHODS AS TOOLS... CHARTING A PATH FORWARD CONCLUSION References

 
Contemporary imaging methods are powerful and are transforming the way research in many disciplines is performed. Consequently, more sharing of the imaging toolkit with investigators outside radiology appears inevitable. If done in the right way, such sharing also appears highly desirable because of the potential synergies offered by interdisciplinary research, where colleagues from inside and outside departments of radiology can leverage each other's core knowledge to do things together that would be much more difficult or impossible to accomplish as individuals.

How will this increasing interest in the use of imaging methods for basic biomedical research affect departments of radiology? First, researchers outside departments of radiology will want and need more and more access to imaging methods. As this happens, some of the same kinds of anxieties prevalent in clinical turf battles will be unavoidable and will need to be addressed institutionally. Who should control core imaging research facilities? How should access be managed for expensive resources like MR imaging devices? How should institutions invest? What is the best governance model?

It is unlikely that any single formula or strategy for answering these questions will work for every academic center; local history is always a major determinant and there are simply too many other variables—including personalities—to posit a single model. However, one compelling observation points to a model where departments of radiology should be asked to host imaging core facilities. This observation hinges on the power of working in interdisciplinary teams that are capable of both furthering the state of the art of imaging and of using the most cutting-edge imaging methods for basic science applications. Investigators from radiology have the expertise in imaging technology—physics, engineering, image capture, and image manipulation and analysis. Investigators from other disciplines bring expertise in their areas of interest—diseases and conditions, molecular and cell biology, and genomics. Working together, an interdisciplinary team gets the best of both. Working alone, nonimaging scientists will find their imaging methods eclipsed and rapidly outdated; they will be less competitive over time without the revitalizing input of imaging colleagues. It is also simply impractical and wasteful to accept that multiple-applications groups in an institution duplicate the same imaging resources in multiple laboratories. Nor can each laboratory expect to maintain a critical mass of talent and equipment in the basic imaging sciences to support their respective imaging efforts.

When interdisciplinary teams are functioning optimally, the scientific process becomes cyclic and reinforcing and not just a linear set of steps from hypothesis to result. As imaging methods are applied, their limitations are discovered and inform the development of better methods. As these new methods come on line, they support, in turn, the asking of more interesting and rigorous research questions. Only an interdisciplinary team is in a position to promote and take advantage of this kind of interplay. Many institutions around the country with successful radiology research programs have adopted this model, whether by prospective design or because of their own empirical experiences.

There are implications of the open-access core facility model for both institutions and departments. Institutions must be educated to invest in them. They need to understand that imaging is now one of the underpinnings of basic biomedical research. Therefore, it is not a question of what the department of radiology "wants" but a question of what investments and initiatives the institution should make to support its overall research establishment. This obviously must be convolved with the spectrum of institutional research interests to determine the nature of facilities required. Small-animal imaging cores, molecular imaging cores, PET facilities, and functional MR imaging laboratories are obvious candidates, depending on institutional interests and priorities.

From a departmental perspective, the challenges include how to balance the research interests of radiology scientists with the service expectations placed on them, how to promote a perception of fairness in access to the core imaging facilities, and how to ensure that the basic premise of the model is, in fact, working—that the core imaging facilities are cutting edge. These are worthwhile challenges to accept for the opportunity to establish a central role in institutional research life.

Departments of radiology may need to expand their basic science faculties to take on the management of institutional core facilities. The economics of this must be negotiated carefully because it is unreasonable to expect someone to support themselves entirely on "soft" grant money while taking on major management responsibilities and mentoring nonimagers in the use of complex systems.

The perception of fair access is critical if the host discipline is to be regarded as a good steward of institutional resources, and limited access can be a source of tension between investigators. At the Massachusetts General Hospital, we established a multidisciplinary oversight group to evaluate proposals for functional MR imaging facility use to help address this point. One of the biggest problems we have seen in managing access is how to diplomatically tell investigators—especially luminaries—that their goals are impractical in situations where they do not understand the limitations of various imaging methods. People see spectacular results and may make unwarranted assumptions about what can be done. By the same token, one of the greatest thrills in working across disciplines is converging on a set of achievable hypotheses and methods and seeing breakthrough results.

	CONCLUSION

TOP INTRODUCTION BASIC RESEARCH AIMED AT... AN ACADEMIC CONUNDRUM IMAGING METHODS AS TOOLS... CHARTING A PATH FORWARD CONCLUSION References

 
For 100 years, radiology pursued its own unique and mostly inward-looking research interests aimed at producing ever better methods. The stunning developments of the past 20 years, especially in functional and molecular imaging, have now propelled radiology to the forefront of cutting-edge biomedical research and have put radiologists and imaging scientists in the position of being key players on interdisciplinary teams addressing the very most interesting and challenging research questions of today.

The full realization of the implications of this stunning metamorphosis in imaging research requires more oars in the water. Departments of radiology are the natural reservoirs of people with the knowledge and talents necessary to exploit imaging more fully in its new research role. Academic departments of radiology should work institutionally to acquire the necessary resources and to alert research leaders outside radiology of new opportunities for interdisciplinary collaboration and synergy.

	FOOTNOTES

 
Author stated no financial relationship to disclose.

	References

TOP INTRODUCTION BASIC RESEARCH AIMED AT... AN ACADEMIC CONUNDRUM IMAGING METHODS AS TOOLS... CHARTING A PATH FORWARD CONCLUSION References

 

1. Thrall JH. American College of Radiology primer on molecular imaging. J Am Coll Radiol 2004;1(1S):1–32.
2. Ntziachristos V, Bremer C, Weissleder R. Fluorescence imaging with near-infrared light: new technological advances that enable in vivo molecular imaging. Eur Radiol 2003;13:195–208.[Medline]
3. Carson PL, Giger M, Welch MJ, et al. Biomedical Imaging Research Opportunities Workshop: report and recommendations. Radiology 2003;229:328–339.[Free Full Text]
4. Partain CL, Gazelle GS. Biomedical Imaging Research Opportunities Workshop II: report and recommendations. Radiology 2005;236:389–403.[Free Full Text]
5. Kalra MK, Maher MM, Rizzo S, Saini S. Radiation exposure and projected risks with multidetector-row computed tomography scanning: clinical strategies and technologic developments for dose reduction. J Comput Assist Tomogr 2004;28(suppl 1):S46–S49.[CrossRef][Medline]
6. Kalra MK, Rizzo SM, Novelline RA. Reducing the radiation dose in emergency computed tomography with automatic exposure control techniques. Emerg Radiol 2005;11:267–274.[CrossRef][Medline]
7. Hendee WR, Chien S, Maynard CD, Dean DJ. The National Institute of Biomedical Imaging and Bioengineering: history, status and potential impact. Radiology 2002;222:12–18.[Abstract/Free Full Text]
8. Belliveau JW, Kennedy DN, KcKinstry RC, et al. Functional mapping of the human visual cortex by magnetic resonance imaging. Science 1991;254:716–719.[Abstract/Free Full Text]
9. Thrall JH. Reinventing radiology in the digital age. I. The all-digital department. Radiology 2005;236:382–385.
10. Strauss MM, Makris N, Aharon I, et al. fMRI sensitization to angry faces. Neuroimage 2005;26:389–413.[CrossRef][Medline]
11. Thermenos HW, Seidman LJ, Breiter H, et al. Functional magnetic resonance imaging during auditory verbal working memory in nonpsychotic relatives of persons with schizophrenia: a pilot study. Biol Psychiatry 2004;55:490–500.[CrossRef][Medline]

This article has been cited by other articles:

				C. D. Maynard Eugene W. Caldwell Lecture 2007: Radiology Research Good to Great? Am. J. Roentgenol., October 1, 2007; 189(4): 757 - 764. [Full Text] [PDF]

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Thrall, J. H. Search for Related Content PubMed PubMed Citation Articles by Thrall, J. H.