IUPS Physiome Project
Presented to the IUPS Council at the 32nd World Congress in 1993, the physiome project is intended to provide a quantitative description of physiological dynamics and functional behavior of the intact organism. The Physiome Project was designed and developed over the next two World Congresses, and in 2001, it was designated by the IUPS executive as a major focus for IUPS during the next decade. Professor Peter Hunter was appointed Chair of the newly created Physiome Commission of the IUPS in 2000.
The Physiome Project aims to explain how each and every component in the body works as part of the integrated whole. Major diseases like cancer and neurological and cardiovascular diseases are complex in nature, involving everything from genes to environment, lifestyle and aging. Integrating knowledge of all these different components into robust, reliable computer models will yield enormous medical advances in the shape of new therapies and diagnostic tools. Ultimately, the goal of the VPH/Physiome Project is to piece together the complete virtual physiological human: a personalized, 3-D model of an individual’s unique physiological make-up. Clinicians will use virtual individuals for applications such as trialing drugs, personalizing medicine (including designing implants to suit a particular person’s body) and performing virtual ‘surgery’ to gauge the outcome of a proposed operation.
The Auckland Bioengineering Institute, as a key player in the VPH/Physiome Project, has led the development of mark-up languages for encoding anatomical and physiological models that have become the mandated standards for European Commission and National Institutes of Health- funded research; set up standardized model repositories that researchers use worldwide; and developed many of the online tools for authoring, visualizing, executing and analyzing the models that the Auckland Bioengineering Institute and other research institutions build. Bringing mathematics, physics, computing and engineering to bear on biology can produce new understandings on how a ‘parts list’ for life can integrate into the whole body.
The VPH/Physiome Project dates back to informal beginnings in the early 1980s when Professor Peter Hunter and Professor Bruce Smaill (now the Director and Deputy Director of the Auckland Bioengineering Institute respectively) started modeling the heart. The Physiome Project was not officially launched until 1997 when the International Union of Physiological Sciences (IUPS) set up a committee headed by Professor Hunter to govern the development of models for all 12 organ systems in the body.
The work of the IUPS Physiome Project has been boosted by the European Commission-funded Virtual Physiological Human Initiative. The VPH Initiative’s main goals are to develop patient- specific computer models for personalized healthcare and simulate disease-related processes. The initiative comprises more than 20 projects, including for example euHeart, the development of personalized heart models for diagnosing and treating heart disease; and HAMAM, highly accurate breast cancer diagnosis technology that combines novel imaging methods with modeling.
Developing a modelling framework
An important aim of the Physiome Project is to develop a multi-scale modelling framework for understanding physiological function that allows models to be combined and linked in a hierarchical fashion. Electromechanical models of the heart, for example, need to combine models of ion channels, myofilament mechanics and signal transduction pathways at the subcellular level and then link these processes to models of tissue mechanics, wavefront propagation and coronary blood flow – each of which may well have been developed by a different group of researchers.
Models can be defined at various levels of abstraction:
1. The conceptual level (the domain of a biologist) - words are used to describe the model;
2. The mathematical level (the domain of the bioengineer and applied mathematician) - the domains, fields, equations and boundary conditions are defined in standard mathematical notation;
3. The formulation level (the domain of the mathematical modeler) – the equations are formulated in terms of the solution method e.g. FEM (i.e. domains, fields and boundaries are described in terms of meshes and parameters);
4. The solution level (the domain of the numerical analyst) – involves the algorithms for solving the parameterised equations on the parameterised domains.
To facilitate communication between research groups and across these levels of abstraction, markup languages are needed to encapsulate the mathematical statements of the governing equations (MathML) and the way in which spatially and temporally dependent continuum fields are parameterised (FieldML). Scripting languages (such as Perl, Python or Matlab) are then needed to create modules which implement the mathematical operations on these fields and call libraries of numerical solution algorithms.
The framework which includes MathML, FieldML, the scripting modules and a grouping construct (components, variables, imports etc – see below) is called ‘ModelML’. When domain specific ontologies (controlled vocabularies together with domain specific rules) are added to link this framework to biological entities such as cells, tissues or organs, the markup language framework is extended to become TissueML or AnatML.
In Cmgui, computed variables are used to describe mathematical operations (such as spatial differential operators) on existing fields using algebraic operations rather than numerical differencing. Computed variables will be used to construct the functionals that need to be minimised at the formulation level. It is intended that these computed variables will be constructed from MathML.