Biomedical Engineering Reference
In-Depth Information
In addition to computational challenges, there are also many conceptual chal-
lenges. For example, events at the tissue level may depend on past events and take
hours to days to occur, while at the intracellular level processes may occur in a
fraction of a second. In addition to integrating across temporal scales, integrating
across spatial scales (i.e., 2D versus 3D) may require further dispensation. One of
the major challenges remains in simplifying the complex system due to gaps in our
understanding or in an attempt to not over constrain the model. Where should one
start, and how is each decision justified? Another challenge of integrating multiple
discrete models is deciding what information to pass back and forth. Each model
may have interdependency within itself, thus passing a concentration or stress
value negates any feedback mechanisms the model had related to these parameters.
Therefore, integrating models that rely on values from one another is a challenge.
Biological adaptation and variability are difficult to capture in a universal
mathematical model. How biological systems change in time is what the models
presented herein try to account for, but some adaptations are unpredictable. For
example, natural effects (due to ageing, hormonal life cycles, ones genetic
makeup, even what division cycle cells in the body are on) and external effects
(due to accidents, smoking, exercise, eating habits, radiation, etc.) may alter how
the general process works. Therefore, if the response of one patient or system
could be very different than another, are the models unique to the patient? How
general should the models be? Of course we are currently limited by our tech-
nology to measure and characterize the interactions of phenomenon of biological
systems. Generally speaking, like the Heisenberg uncertainty principle, to augment
our knowledge of, say, the rate of growth factor production may come at the cost
of compromising physiological conditions.
Nevertheless, we feel that complex system modeling in biology is the key to
developing new drugs and therapies over the next 50 years; as such there are
educational needs that should be met. Having more undergraduate courses that
deal with complex systems analysis in biology will equip more students with the
fundamental skills. More graduate courses on the theory of modeling vascular
adaptation, and biological adaptation in general, will allow for specialization and
additional improvements. Having more graduate programs and/or cross-degree or
dual-degree Ph.D. programs that are designed to treat high-throughput data in the
context of in vivo function and quantitative modeling is needed. In addition,
continued changes in academic culture that recognize the value of collaboration
and teamwork on large complex systems will facilitate more advanced models. We
are encouraged to hear that in April 2012, NSF and NIH jointly launched a ''Core
Techniques and Technologies for Advancing Big Data Science and Engineering
(BIGDATA)'' initiative. The need for a means to manage, analyze, visualize, and
extract data from diverse, distributed data sets has been recognized. If successful,
having this wealth of ordered data at our fingertips will only help to update and
improve the rules and relations of multiscale modeling.
Acknowledgments
This work was supported, in part, via NIH grants HL-86418 to JDH and HL-
82838 to SMP.
Search WWH ::
Custom Search