Using semantic dependencies for consistency management of an ontology of brain–cortex anatomy

Abstract

ContextA symbolic model allows developers to represent general knowledge about a domain and the meaning that is commonly associated with it. This knowledge can be used by itself (e.g., for teaching), or indirectly as a reference to process specific facts (e.g., to assist queries or data retrieval). In the latter case, symbolic models are perceived as a key feature to provide software assistance for tasks that now require domain-aware intervention by a human. As interoperability of these software applications is desirable, shared conceptual models, and specifically ontologies, play a major role in a Semantic Web context [1]. Since these models are to be usable by software, they must meet explicit representation and consistency requirements. For medical applications, anatomy provides a common reference used to reason about pathology or localization of functional activity [2,3]. The Foundation Model of Anatomy (FMA) [4] and Galen [5] are two major conceptual models that provide a symbolic representation of human anatomy. However, neither of them provides a satisfactory representation of brain-cortex anatomy. The major sources of neuroanatomical knowledge are paper-based atlases [6,7] and terminological systems such as Neuronames [8].We are working on an ontology of brain-cortex anatomy. Our goal is more to formalize existing knowledge than it is to propose new anatomical concepts or relationships. Our model has been described in previous publications [9,10]. It comprises 304 concepts and 1254 relationships that represent the organization of anatomical structures. Because the brain surface presents complicated folding patterns, typical anatomical structures are gyri (the bulges of cerebral matter, similar to hills), the sulci (the hollow foldings, similar to valleys) and lobes (sets of gyri). Our model's taxonomy hierarchy is composed of three levels. First, the generic level contains concepts such as Lobe or Sulcus, and is mainly used to define the domain and range of the relationships. Second, the abstract level represents a prototypical brain hemisphere, and contains concepts such as Frontal Lobe or Central Sulcus. Third, the lateralized level is used to represent left/right asymmetries, and contains concepts such as Left Frontal lobe. For mereology, the model identifies several relationships such as hasDirectAnatomicalPart, hasAnatomicalPart, hasSegment and their properties, inspired from previous theoretical works [11,12]. The model also represents neighborhood relationships such as the separation of two cortical structures by a sulcus, anatomical continuity, and sulci connection.In this context, managing the semantic consistency of the ontology has been one of our main concerns.

An analysis of FMA using structural self-bisimilarity

“…Third, we not only discover those concepts with high likelihood of errors, but also point out the roots where the errors were introduced. The eight patterns for the problematic Matches also help the domain experts trace back to the original conceptual errors, which can be useful in discovering and creating new design patterns in the processes of creating new ontologies [1, 24, 25]. …”

“…The eight patterns we presented above provide the human editors of biomedical ontologies a full dimensional view of all the structures for Matches, which can be helpful in discovering and introducing new ontology design patterns [24, 25]. …”

“…In all, we conduct a scalable, comprehensive analysis of all the possible structures for each symmetric pair, which is not just for detecting potential errors, for analyzing those scenarios, but more importantly, for exposing the sources of errors, and for guiding the creation of new ontologies with potential ontology design patterns [24, 25]. …”

Mapping the Human Brain: New Insights from fMRI Data Sharing

“…This "meta-data" is the important contextual information needed to reproduce the results of the study with maximal fidelity. Several approaches to standardize meta-data organization and format have been discussed elsewhere (Kotter and Wanke 2005;Dameron and Musen 2007) and an elaboration of their merits is beyond the scope of this particular article. It suffices to say that carefully constructed frameworks for meta-data description that promote efficient data exchange and re-use need to be developed and put in practice.…”

Generation and Use of One Ontology for Intelligent Information Retrieval From Electronic Record Histories

“…Our third objective is to confirm the hypothesis that the constituent elements of each semantic class represent a perfectly determinable and controllable set. A class may have superclasses which offer a medical meaning membership from one given concept to another with a more general meaning (Dameron, 2004).…”