Proceedings of the 26th Annual International Conference of the IEEE EMBS

San Francisco, CA, USA • September 1-5, 2004

The Importance of Multidisciplinary Teams

In a Large Biomedical Research Program
R. L. Martino
Division of Computational Bioscience, Center for Information Technology
National Institutes of Health, Department of Health and Human Services, Bethesda, MD, USA

teams was considered to be so important a topic that it was

Abstract—The formation of multidisciplinary teams is vital to
modern biomedical research. These teams are important
contributors to the advancement of scientific discovery as well
as the translation of those discoveries into useful clinical
practice. The issues involved in providing scientific and
technical expertise in computational science and engineering to
support a large biomedical research program are presented.
The activities of the Division of Computational Bioscience
(DCB) of the Center for Information Technology (CIT) at the
National Institutes of Health (NIH) are highlighted.
Professionals in this organization apply the concepts and
technologies of computer science, engineering, physical science
and mathematics to biomedical applications, making available
this expertise to the NIH Intramural Research Program (IRP).
The areas of application include imaging, informatics,
instrumentation, telemedicine, structural biology and
mathematical analysis. Examples where collaborative research
and development teams containing DCB professionals have
worked to produce significant results are presented. These
examples are a chromosome microdissection instrument, a
telemedicine system for radiation oncology, and a medical
image archive.
Keywords— Biomedical imaging, biomedical
instrumentation, collaborative research, multidisciplinary
teams, telemedicine
I. INTRODUCTION
made one of the roadmap themes. The stated motivation for
this theme is that the scale and complexity of today’s
biomedical research problems increasingly demands that
scientists move beyond the confines of their own discipline
and explore new organizational models for team science.
A subcommittee of the second NIH Roadmap theme,
Teams of the Future, looked at the NIH Intramural Research
Program (IRP) as a model for multidisciplinary and
interdisciplinary research [3]. The subcommittee sent a
questionnaire to all NIH Scientific Directors to determine
the best practices associated with facilitating research within
and among the NIH Institutes and Centers. This information
was synthesized to provide a description of the perceived
barriers and the lessons learned related to multidisciplinary
and interdisciplinary research. Although a number of
barriers were identified, it was discovered that remarkably
few insurmountable barriers exist. Several common lessons
were learned including the following issues that should be
addressed when establishing a multidisciplinary or
interdisciplinary research team: clearly articulate the mission
of the team, carefully determine membership, provide team
leadership, implement clear methods of communication,
provide training and education, supply adequate funding,
create a positive environment, and insure appropriate review
and reward.

The formation of multidisciplinary teams is vital to II. METHODOLOGY

modern biomedical research. These teams are important
contributors to the advancement of scientific discovery as
well as the translation of those discoveries into useful
clinical practice. In order to conduct complex and large-
scale biomedical research, the expertise of many disciplines
are often required and combined to create a synergistic
entity that can solve the problems encountered during a
particular scientific or clinical effort [1].
During the past year, the NIH and its leadership, with
input from recognized leaders in academia, industry,
government, and the public, developed the NIH Roadmap
[2]. This Roadmap is an integrated vision to deepen the
understanding of biology, stimulate interdisciplinary
research teams, and reshape clinical research to accelerate
medical discovery and improve people’s health. The three
NIH Roadmap themes are New Pathways to Discovery,
Research Teams of the Future, and Re-engineering the
Clinical Research Enterprise. In crafting this new vision for
biomedical research, the formation of new types of research
A. Collaborative Research and Development
Vannevar Bush formulated the linear model of
technological development in his 1945 report entitled
Science – The Endless Frontier, a document that governed
the relationship between science and the federal government
until the end of the Cold War [4]. The report promised that
science could provide for national security, national health
and national prosperity if the scientific community received
all of the required resources. In this model, science is
assumed to drive technological innovation in a linear
manner with scientific advances fueling applied research,
then development work, and ultimately produce
commercialization.
Donald Stokes presented major limitations to this long
accepted linear model that are relevant to the discussion of
the present need for multidisciplinary teams in biomedical
research [5]. The linear model ignored fundamental research

0-7803-8439-3/04/$20.00©2004 IEEE 5188

conducted by a researcher inspired by application
considerations. The linear model also assumed a disciplinary
model of academic research not allowing for a
multidisciplinary approach.
To overcome the limitations of the linear model, Stokes
presented the dynamic model shown in Figure 1. This model
shows the interactive relationship between advances in
science and the development of technology. In this model,
the basic science and technological innovation paths can
progress independently. However, each of these paths can be
influenced by the other with application inspired basic
research often serving the linking role. This interactive
model of science and technology demonstrates the
importance that technology advances may have in
supporting new fundamental science and serves as a
framework for conducting a multidisciplinary program.
Collaborative, multidisciplinary teams serve an important
role in the interaction of scientific advancement and
technology development.
B. Forming Multidisciplinary Teams
The Division of Computational Bioscience (DCB) is a
research and development organization that provides
scientific and technical expertise in computational science
and engineering to support the NIH IRP. Working with all
of the Institutes and the Clinical Center, DCB applies the
concepts and technologies of computer science, engineering,
physical science and mathematics to biomedical
applications. The areas of application include imaging,
informatics, instrumentation, telemedicine, structural
biology and mathematical analysis. DCB develops
computational methods and tools for solving biomedical
laboratory and clinical research problems and manages
centralized scientific computational and communication
systems. DCB also provides an environment to mentor
computational scientists and engineers for careers in
biomedical research.
Improved Improved
Understanding Technology
Purely
Application
Pure Applied
Inspired
Basic Research
Basic
Research and
Research
Development
Existing Existing
Understanding Technology
Fig. 1. The Dynamic Model of Science and Technology
Research and Development [5]
5189
The availability of the DCB professional staff allows for
the formation of project teams containing the combination
of computational science and engineering skills needed to
solve multidisciplinary research problems and develop
unique systems. The core disciplines of the DCB staff are
applied mathematics, biomedical engineering, chemistry,
computer science, computer engineering, electronics
engineering, molecular biology, physics, and statistics. The
DCB collaborative research and development program is
focused on the following activities where the expertise
exists within this organization and a meaningful contribution
can be made to the NIH IRP.
Biomedical Imaging: Develop and apply advanced imaging
methods and algorithms to images of biomedical objects that
range in size from the whole body to macromolecules.
Computational techniques are applied to a variety of
biomedical imaging modalities including electron
microscopy, light microscopy, ultrasound, Computed
Tomography (CT), Positron Emission Tomography (PET),
Magnetic Resonance Imaging (MRI), and Nuclear Magnetic
Resonance (NMR) spectroscopy.
Biomedical Informatics: Develop integrated methods for the
analysis, viewing, storage, and management of microarray
data. Develop new tools for integrating information from
genomic and biological knowledge bases and for the
analysis and presentation of gene annotations. Design image
archive and clinical information systems.
Computational Methods and Algorithms, and Parallel
Computing: Develop high-performance computational
methods and algorithms to analyze biomedical research data
and model complex biological systems that may require
multiple processor systems. Implement knowledge-based
data management systems that provide the environment for
collaborative biomedical research studies.
Telemedicine: Implement advanced telemedicine systems
using high-speed communication networks and multi-
modality imaging technologies.
Biomedical Instrumentation: Design unique, real-time
biomedical signal processing and control systems as well as
advanced biomedical instrumentation.
Molecular Modeling: Assemble, develop, and promote
computational tools and expertise for simulating molecular
systems. Collaborate in projects involving molecular
modeling and dynamics, structure prediction, and quantum
chemistry.
Mathematical and Statistical Analysis: Provide
mathematical and statistical expertise to the NIH IRP.
Develop and apply mathematical theories, which describe
physical, chemical and biological processes.
 III. RESULTS To ISDN
Connection
Three examples where collaborative research and
Microphone Audio Speakers
development teams containing DCB professionals have Audio
worked to produce significant results are presented in this S-Video
Tandberg S-Video
section. CODEC
Video Camera

A. Biomedical Instrumentation RS-232

ISDN
Cisco
Router Color Video Monitor
In collaboration with scientists and engineers of the Video
RGB
Ethernet
National Human Genome Research Institute (NHGRI) and Video
the NIH Office of the Director, Office of Research Services
Division of Bioengineering and Physical Science, the DCB
designed and implemented the chromosome microdissection Color Display
Computer
system. Chromosome microdissection is a technique that
physically removes a large section of DNA from a complete
chromosome, making the DNA available for further study. Keyboard
Electronic View Boxes
The smallest portion of DNA that can be isolated using this Mouse or Trackball

method comprises ten million base pairs, hundreds of

Fig. 3. Diagram of the Medical Consultation Workstation
thousands of individual genes. This facilitates the research
of diseases associated with specific chromosomal
B. Telemedicine Systems
abnormalities, which could not be resolved by classical
banding [6].
In collaboration with the medical staff of the National
The chromosome microdissection system automates a
Cancer Institute (NCI) Radiation Oncology Sciences
large portion of the dissection process resulting in higher
Program, the DCB developed the TELESYNERGY®
throughput over manual methods, higher resolution for the
Medical Consultation WorkStation (MCWS) [7,8]. The
acquisition of smaller chromosome fragments, and shorter
TELESYNERGY® System is a multi-media, medical
training periods for users. The system accomplishes these
imaging workstation, used within an electronic imaging
advances by computer control of the cutting needle and the
environment, and utilizing high-speed networking
laser targeting for ablation, laser ablation of chromosome
technologies to establish international links. A diagram of
material neighboring the segment of interest, and point and
the ISDN version of the MCWS is shown in Fig. 1.
click type user interactions instead of manual micro-
The TELESYNERGY® System allows face-to-face
manipulations on the microscope. With the ability to burn
consultation among multiple participant sites, while
away unwanted chromosome material neighboring the
transmitting simultaneous high-resolution images to all sites.
segment of interest, a scientist can obtain a smaller
It also provides a high-quality teleconferencing facility,
dissection area, reduce contamination, and eliminate the
which can transmit information and images from several
need to cut perpendicular to the chromosome. A picture of
different sources such as a document camera, video
the system needle picking up chromosome material is shown
recorder, microscope camera and personal computer display.
in Fig. 2.
In addition to standard teleconferencing, diagnostic-
quality radiology and pathology images can be transmitted
and viewed. Each of the two high-resolution monochrome
image displays function as an Electronic View Box for the
presentation of full-size (14 X 17-inch format) "electronic
films.” Utilizing these displays, users can perform organ and
lesion contouring via a shared-cursor technique, in
consultation mode, which allows clinicians to collaborate in
feature identification. The identified regions-of-interest are
viewed simultaneously and in real-time between all
participating TELESYNERGY® sites. A remote-controlled
Olympus microscope with a video camera and remote
control module allows users to manipulate and assess biopsy
specimens from a number of sites concurrently. The initial
clinical application area for TELESYNERGY® was in
support of radiation treatment planning.
Fig. 2. Picture of Needle Picking Up Dissected Chromosome Material

5190
C. Biomedical Informatics
Biomedical research at the National Institutes of Health
(NIH) involves the acquisition, storage, communication,
display, and analysis of large numbers of medical images
including many image modalities for both animal and
human studies. NIH researchers use these medical images
to discover biomarkers for disease states, determine the
genetic origin of disease, evaluate the results of clinical
trials, and develop therapeutic procedures. This
accumulation of many medical images for a variety of
research protocols and the desire to access and analyze them
over many years generates the need for effective medical
image archive systems.
In collaboration with the National Institute of
Neurological Disorders and Stroke (NINDS), DCB
developed a Web-based medical image archive system [9].
NINDS requires a central image repository to address
questions of stroke pathophysiology and imaging
biomarkers in stroke clinical trials by analyzing images
obtained from a large number of clinical trials conducted by
government, academic and pharmaceutical industry
researchers.
The Web-based medical image archive system employs
a three-tier, client-server system architecture, incorporating
modern database, secure Web, middleware software,
network storage, and networking technologies. This three-
tier system design gives users secure access to distributed
system resources that appear as a single unified archive
system from their desktop computer through user-friendly
Web-interfaces over the Internet. The design goal was to
develop a scalable and upgradeable central image archive
system to store and disseminate medical image data through
easy Web access over the Internet. The core of the Web-
based medical image archive system is implemented using
SUN Enterprise servers connected over 100 mega-bits per
second Ethernet. Benchmark results on the image archive
system, consisting of a SUN 6500 database server, a SUN
450 application server, and networked RAID storage
systems, show consistently fast system response times as if
uploads, queries, and downloads were performed at the
user’s local desktop computer.
IV. DISCUSSION
As the dynamic model of science and technology
research and development described in the methodology
section demonstrates, technology advances can make
important contributions to progress in fundamental science.
Multidisciplinary teams serve an important role in the
interaction of scientific advancement and technology
development.
The collaborative research activities of the DCB provide
examples of the contributions multidisciplinary teams can
make to advances in instrumentation, telemedicine, imaging,
5191
and informatics. The chromosome microdissection system
provides new avenues of research on diseases associated
with chromosomal abnormalities. The TELESYNERGY®
System is being used for clinical trials, whose outcomes
may lead to innovative new treatment protocols that could
improve the prognosis for future cancer patients. The Web-
based medical image archive system gives NINDS
researchers and clinicians easy access to stroke MRI and CT
images, as well as clinical trial data, for their research in
developing experimental stroke therapies using imaging
surrogate markers.
ACKNOWLEDGMENT
The author would like to gratefully acknowledge the
contributions of all the members of the Division of
Computational Bioscience of the Center for Information
Technology at the NIH including H. Cheung, C. A. Johnson,
K. M. Kempner, P. J. Munson, T. J. Pohida, J. I. Powell, P.
J. Steinbach, and B. L. Trus.
REFERENCES
[1] S. J. Nass and B. W. Stillman, Editors, Large-Scale Biomedical
Science, Exploring Strategies for Future Research, Washington,
D.C.: The National Academies Press, 2003.
[2] E. Zerhouni, “The NIH Roadmap,” Science, vol. 302, pp. 63-64,
72, 2003.
[3] J. C. Barrett, Report From the Subcommittee on Using the NIH
Intramural Research Program as a Model For Interdisciplinary
Research, Bethesda, Maryland: National Institutes of Health,
2003.
[4] C. Tucker, “Public Policy Towards Science and Technology,” in
Wiley Encyclopedia of Electrical and Electronics Engineering,
J.G. Webster, Ed. New York: Wiley and Sons, Inc., 1999, vol.
17, pp. 437-446.
[5] D. E. Stokes, Pasteur’s Quadrant, Basic Science and
Technological Innovation, Washington, D.C.: Brookings
Institution Press, 1997.
[6] J. Zhang, J. M. Trent, and P. S. Meltzer, “Rapid Isolation and
Characterization of Amplified DNA By Chromosome
Microdissection: Identification of IGF1R Amplification in
Malignant Melanoma,” Oncogene, vol. 8, pp. 2827-2831.
[7] K. M. Kempner, F. S. Govern, and R. L. Martino, “Telemedicine
Systems,” in Introduction to the Principles and Practice of
Clinical Research, J.I. Gallin, Ed. New York: Academic Press,
2002, ch. 22, pp. 361-385.
[8] R. L. Martino, K. M. Kempner, F. S. Govern, D. Chow, M. E.
Steele, J. E. Elson, and C. N. Coleman, “A Collaborative
Telemedicine Environment For the Ireland - Northern Ireland -
National Cancer Institute International Partnership in Cancer
Care,” Proceedings of the 25th Annual International Conference
of the IEEE Engineering in Medicine and Biology Society,
Piscataway, New Jersey: IEEE Press, 2003, pp. 1350-1353.
[9] Suh, E.B., S. Warach, H. Cheung, S.A. Wang, P. Tangiral, M.
Luby, and R.L. Martino, “A Web-based Medical Image Archive
System,” in E.L. Siegel and H.K. Huang, editors, Medical
Imaging 2002: PACS and Integrated Medical Information
Systems: Design and Evaluation, Proceedings of the Society of
Photo-Optical Instrumentation Engineers, Bellingham,
Washington: The International Society for Optical Engineering,
vol. 4685, pp. 31-41, 2002.