Toward Useful Biomedical Models
Dr. Grace C.Y. Peng
National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health
Bethesda, MD, USA
In the transportation industry, computational models are integrated in virtually every aspect of the vehicle design, safety and performance testing; in fact it would be impossible for a traveler to choose to fly in an airplane that was designed without the use of a model. On the other hand, in medical practice few models are used for clinical decision support or personalized drug dosing; yet patients accept this as the standard of care. In basic biomedical research more and more models are used to validate data, but few models are actually used to drive the design of experiments or develop testable hypotheses. In our modern day of ubiquitous computing, it is curious why this dichotomy exists between the engineering and biological worlds.
Nevertheless this is an exciting time for the field of modeling of biomedical systems. In 2003, a small trans-agency working group in the U.S. government resulted in the formation and growth of the Interagency Modeling and Analysis Group (IMAG) and subsequently the release of the first interagency solicitation for multiscale modeling of biomedical, biological and behavioral systems. That solicitation funded 24 projects, creating the Multiscale Modeling Consortium (MSM). The IMAG MSM Consortium is now in its fifth year. During this time many other multiscale modeling initiatives have emerged from the now eight IMAG government agencies. In addition to multiscale modeling IMAG is also exploring population models that approach modeling from the phenomenological point of view and the possibilities for leveraging these models with the mechanistic models developed by MSM participants. Furthermore, there is a swelling interest from the international community to coordinate efforts for multiscale, physiome modeling. In order to move forward with these modeling efforts, it is therefore of utmost importance to address how models can indeed impact the broader communities and clearly understand the limitations in making this impact occur.
On December 15-16, 2009 IMAG held the first IMAG Futures Meeting - The Impact of Modeling on Biomedical Research. This meeting was an opportunity for IMAG to assess to what extent computational modeling has succeeded or failed to make a difference in the broader research endeavor, and to discuss these issues in the context of current challenges and opportunities for biomedical, biological and behavioral modeling. This was a brainstorming meeting that included government and community leaders, as well as attendees interacting via worldwide videocast. The discussions were grouped by five biological scales: 1) Population; 2) Whole-Body; 3) Cell-Tissue-Organ; 4) Pathways and Networks; and 5) Atomic and Molecular. Participants were encouraged to reflect on their own fields and expertise and call attention to issues that are unique to each scale as well as issues that may span across scales. Pre-meeting discussions and an open post-meeting public commentary period resulted in an IMAG Futures report. In this talk, I will summarize the findings of this report and highlight aspects that are especially pertinent to high performance computing applications in biomedical research and policy.
For more information on IMAG and MSM activities, please visit the IMAG
wiki, http://www.imagwiki.org/mediawiki.
|
