GFO-Based Patterns for Situation Modeling: From the Cell Tracking Ontology to Patterns Applicable in the Biomedical Domain

Abstract

The concept of situation is crucial for many domain ontologies, particularly in complex fields such as biomedicine. For instance, work on the Cell Tracking Ontology (CTO) revealed that the notion of situation is fundamental for the adequate representation of cellular genealogies, which provide a detailed record of the lineage and development of cells over time. Related findings were observed in other biomedical use cases, such as surgical navigation, medical risk identification, and pregnancy care and complication prevention. The development of these applications heavily rests on the General Formal Ontology (GFO, version 1.0), which effectively represents situations. However, different terminologies and modeling variations have been used across these applications, leading to modifications and extensions of GFO. We aim to generalize the results from CTO and other GFO applications towards a systematic basis for situation modeling. The paper reviews four biomedical projects using GFO for situation modeling and discusses the GFO theory of situations, along with its underlying intuitions. Based on this, we introduce a family of ontology design patterns tailored to modeling situations, which can be challenging, as an optimal approach often depends on the specific use case and requirements. However, the range of approaches observed in real-world scenarios is sufficiently limited to allow for a systematic classification of the modeling methods, addressing various levels of detail. Therefore, the family of patterns aims at providing a toolbox for depicting situations at different granularities, including their participants, qualities, dynamics, and changes over time, rather than forcing a one-size-fits-all approach.