Editors' ChoiceComputational Medicine

The computational model will see you now

See allHide authors and affiliations

Science Translational Medicine  09 Sep 2015:
Vol. 7, Issue 304, pp. 304ec154
DOI: 10.1126/scitranslmed.aad3069

With computational power increasing, mathematical equations quantifying human physiology have been connected across biological time and length scales to develop multiscale models, from the subcellular level all the way to the organ and organism levels. These models can be used to expose critical gaps in our knowledge of human biology and disease and also motivate simulations of how a disturbance on one scale manifests at the systems level. Given the explosion in available biomedical and healthcare data, it is natural to ask whether such models can be personalized. Vicario and coworkers work toward this goal, taking a different approach to personalized physiological modeling and parameter estimation through “minimal modeling”—reducing the model to a few crucial parameters (aggregation).

Complex models with long simulation times and thousands of model parameters may not be compatible with personalized medicine. Rather than starting with multiscale models, Vicario et al. applied a highly aggregate representation of the human respiratory system to estimate lung resistance and elastance from select data from spontaneously breathing, ventilated patients. By incorporating reasonable physiological constraints into their optimization algorithm, the authors were able to accurately estimate resistance and elastance in both simulations and in a large animal (pig) study where gold-standard invasive measurements had also been conducted. The algorithm can be run in real time, thus potentially aiding critical care at the bedside. An open question is how one can bridge small-scale, aggregate models (such as in Vicario et al.) and large multiscale models, linking personalized outputs to detailed mathematical representations of human physiology. If such personalized models can be realized, they would have the potential to help guide patient care, particularly in data-rich clinical environments, such as critical care medicine.

F. Vicario et al., Noninvasive estimation of respiratory mechanics in spontaneously breathing ventilated patients: A constrained optimization approach. IEEE Trans. Biomed. Eng. 10.1109/TBME.2015.2470641 (2015). [Abstract]

Stay Connected to Science Translational Medicine

Navigate This Article