Abstract

The creation of a conceptual schema that symbolically represents the collective knowledge of a particular domain is a daunting task, particularly to capture and unify the breadth and complexity of human form and function. Much like other modelling endeavours, a number of fundamental challenges exist when developing a formal knowledge base. The primary goal here is to identify the key elements that must be considered during ontology creation and evolution, so that we can facilitate the desired compilation and sharing of multi-modal, multi-dimensional anatomical data. What is required to accommodate a growing, iteratively-authored, and searchable data-driven knowledge base? How do we annotate and compile multiple instances of anatomical structures into a probabilistic framework that accounts for inter-individual variability, including the potential existence of geometrical asymmetries, anomalies, abnormalities and pathologies? A lack of common standards can result in massive information loss, both in terms of what is (or can be) analytically gleaned from the data, as well as the contextual information that guides observations and analyses. Thus, we strive to develop a flexible ontology that incorporates sufficient granularity to support the assembly of a "complete" human model that enables multi-purpose, multi-scale modelling and simulation.