James Matthew Fielding’s research while affiliated with Paris 1 Panthéon-Sorbonne University and other places

We would first of all like to thank Thor Grünbaum and Andrea Raballo for their thoughtful and lively commentary on our work. We would also like to thank Daniel Rubin for taking this opportunity to describe in detail some of the research carried out in this domain since our paper was first written. Although their commentaries may seem to fall on opposite ends of the critical scale, so to speak, taken together they provide an opportunity to take stock of the progress that has been made in this endeavor as well as some of the challenges that remain to be overcome. In this short response, we focus primarily on what we believe to be the most serious of Grünbaum and Raballo's criticisms to our work, arguing that our position is far more modest than their characterization suggests and is indeed rather closer to the one they themselves claim to be entirely unobjectionable. Our aim here is to argue that the advances to be made in this endeavor—even if they are unobjectionable—are far from trivial as Grünbaum and Raballo claim. Turning to Grünbaum and Raballo's commentary, we believe there are three principle criticisms around which their arguments turn. The first two concern the very possibility of implementing some form of applied ontology in the psychiatric domain, regardless of the form this application would take. Their third criticism concerns the particular ontological framework we have applied, adapted from Roman Ingarden's ontological aesthetics. We first treat Grünbaum and Raballo's general criticisms, followed by the more specific. Where possible, we also turn to Rubin's commentary for further demonstration of our main points. The first of the general criticisms Grünbaum and Raballo put forward is that, although our applied ontology is intended to aid researchers and medical practitioners in the field of neuroimaging, the current state of the art essentially forbids us from making any advances here. According to them, "there is no well-established consensus about what exactly pictures of patterns of brain activation are showing us about the brain and its role in the performance of cognitive tests" (2011, 306). Because of this lack of clear consensus on the relationships between the physical structures of the imaged brain and the nature of the experience these images are intended to provide some measurement of, they suggest such a knowledge base of neuroimages would be unable to gain an ontological foothold. We would respond, however, that if, as they claim, "unlike x-rays of fractured bones, in psychiatry the relation between the brain image and the disease entity is highly problematic" (2011, 306), this is no argument against introducing a data management tool such as the one we have described. It is, on the contrary, only further evidence of its necessity. It is precisely because of this lack of consensus that the vast amount of image data being accrued needs to be readily accessible and stored in a format that is flexible enough to facilitate the wide and evolving interests and uncertainties of the researchers working in this domain, which is today being carried out at an explosive pace. As Rubin notes, "although nonimage data are easily processed by machines, image data are generally not exploited directly—images typically are stored in archives, and only particular data needed for the study of the in which the images were originally acquired are generally available for subsequent analysis" (2011, 311). Addressing this limited access was one of the principle motivations behind the development of our imaging ontology framework. The second general claim Grünbaum and Raballo make is that, in the absence of a general consensus among practitioners, the scope of the general biomedical imagining ontology we outline, were it applied to the psychiatric domain, would be too limited to be useful. Grünbaum and Raballo are absolutely correct when they state that, "one of the key features of a knowledge representation system is that with the right software it should be able to generate sounds and practically valuable inferences from input information" (307). They are also correct when they claim that such an ambition is "highly problematic" in the psychiatric domain; this is...

Biomedical software ontologies provide a means for the representation of facts gathered through biomedical research and clinical observation. At the foundation of good software ontology design lays a sound philosophical realism that supplies the basic framework required to support the computable management of this information correctly and consistently. In numerous biomedical subdomains (such as anatomy, disease classification, or functional genomics), a good degree of success has been achieved through the realist approach. In the field of psychiatry, however, the analytic tools of ontological realism are challenged to account for subjective mental experiences that typically lay beyond their scope. Although psychiatric symptoms, such as delusions, hallucinations, or memory loss, may be too ethereal to account for in terms of a realist ontology, by focusing on some psychiatric signs, such as images of the human brain (which are in themselves subject to ontological analysis), we may be able to make some in-roads toward an application ontology of the psychiatric domain. In this paper, via the ontological framework of Polish phenomenologist Roman Ingarden, we discuss the differences between the ontology of the body and the ontology of the image, and apply the subsequent image-ontology framework to the domain of neuroimaging. We aim to demonstrate how such an ontology may lead to the perspicuous structuring of clinical information in psychiatry and the benefits application ontologies afford may subsequently be attained within a portion of this particularly difficult domain.

Biomedical ontologies define entities and relations in order to represent knowledge in the biomedical domain. In addition, many ontologies further represent supplementary knowledge by linking terms from an external controlled vocabulary to the entities defined within the ontology itself. In this paper we concentrate on the domain of medical imaging, for which controlled vocabularies are emerging, but no ontology currently exists. We analyzed computed tomography reports in order to determine to which entities terms used in such reports refer and which relations are used with regard to the recently published Open Biomedical Ontologies (OBO) Relation Ontology. Our analysis revealed that the majority of entities referred to in radiological reporting practice are anatomical entities and anatomical coordinates. Based on the Open Biomedical Ontologies (OBO) Relation Ontology we provide a set of additional relations for the ontological structuring of image features such as shape, morphology, size and signal, frequently found in the reports. On the basis of these results we conclude that the construction of an imaging ontology may benefit greatly from already existing reference ontologies such as the Foundational Model of Anatomy (FMA), which represent those entities to which radiological reports most frequently refer.

The central hypothesis underlying this communication is that the methodology and conceptual rigor of a philosophically inspired formal ontology can bring significant benefits in the development and maintenance of application ontologies [A. Flett, M. Dos Santos, W. Ceusters, Some Ontology Engineering Procedures and their Supporting Technologies, EKAW2002, 2003]. This hypothesis has been tested in the collaboration between Language and Computing (L&C), a company specializing in software for supporting natural language processing especially in the medical field, and the Institute for Formal Ontology and Medical Information Science (IFOMIS), an academic research institution concerned with the theoretical foundations of ontology. In the course of this collaboration L&C's ontology, LinKBase, which is designed to integrate and support reasoning across a plurality of external databases, has been subjected to a thorough auditing on the basis of the principles underlying IFOMIS's Basic Formal Ontology (BFO) [B. Smith, Basic Formal Ontology, 2002. http://ontology.buffalo.edu/bfo]. The goal is to transform a large terminology-based ontology into one with the ability to support reasoning applications. Our general procedure has been the implementation of a meta-ontological definition space in which the definitions of all the concepts and relations in LinKBase are standardized in the framework of first-order logic. In this paper we describe how this principles-based standardization has led to a greater degree of internal coherence of the LinKBase structure, and how it has facilitated the construction of mappings between external databases using LinKBase as translation hub. We argue that the collaboration here described represents a new phase in the quest to solve the so-called "Tower of Babel" problem of ontology integration [F. Montayne, J. Flanagan, Formal Ontology: The Foundation for Natural Language Processing, 2003. http://www.landcglobal.com/].

In this paper we isolate four diagnostic models in radiology and define a set of diagnostic relations corresponding to each clinical situation. To achieve this, we describe a set of general formal ontological notions, as well as the ontological model of the imaging domain we employed in our analysis. On the basis of our results, we conclude that these diagnostic models and the relations contained therein could be applied to diagnostic situations outside of radiology as well.

Biomedical ontologies define entities and relations in order to represent knowledge in the biomedical domain. In this paper we concentrate on the domain of medical imaging. In previous work, we analyzed a representative sample of computed tomography reports in order to determine to which entities and relations the terms used in such reports refer (with regard to the Foundational Model of Anatomy (FMA) and the recently published Open Biomedical Ontology (OBO) Relation Ontology, respectively) in order to construct an imaging ontology for electronic reporting in radiology. In this paper we expand the role of two OBO relations in particular, as they may be applied to radiological image information: the relations located_in and adjacent_to. Defining these relations in terms of the basic topological relations of Region Connection Calculus (RCC), we show how the qualitative description of image feature locations in radiological reporting may be formalized for reasoning.

PURPOSE The RSNA is developing a lexicon (RadLex) to unify radiological terms and to facilitate indexing and retrieving of radiological information sources. The purpose of this study was to evaluate which quantity of radiological terms found in thoracic computer-tomography (CT) reports could be matched against RadLex terms and to quantify the use of RadLex term categories for the indexing of textual reports. METHOD AND MATERIALS 250 thoracic CT-reports were selected randomly from a database. Terms were extracted from the reports by using a text-processor and revised by two radiologists. Terms considered relevant were matched against terms contained in RadLex. The reports were indexed manually by two radiologists in RadLex term categories of Clinical History, Image Quality, Anatomic Location, Findings, Relationships and Conclusions. Quantitative analysis of terms classified and indexing results are demonstrated. RESULTS 512 distinct terms were found in the reports considered relevant for classification. 348 (68%) of those terms were found in RadLex term categories of Findings (191, 55%) and Anatomic Location (157, 45%). Results of indexing textual reports are: most frequently used categories were the subcategory of Visual Features (3860) and the category of Anatomical Location (1558). Morphological and physiological processes were encoded 648 times, Diagnoses and Etiologies 351 times, respectively. Spatial Relationships were indexed 1350 times, Causal Relationships 242 and Logical Relationships 120 times. Indexing of terms in the category of Clinical History occurred 325 times. Image Quality was reported as diagnostic or exemplary in 239 reports, 152 Conclusions were indexed definitely clinically significant. CONCLUSION RadLex is a valuable resource to index thoracic CT reports as the majority of radiological terms found in the reports are present in the lexicon. The most frequently used term category for indexing reports was the subcategory of Visual Features which emphasizes the descriptive nature of radiological reporting. Future work will include the construction of teaching files from the indexed reports and the identification of typical reporting patterns on a larger sample of reports.

Software application ontologies have the potential to become the keystone in state-of-the-art information management techniques. It is expected that these ontologies will support the sort of reasoning power required to navigate large and complex terminologies correctly and efficiently. Yet, there is one problem in particular that continues to stand in our way. As these terminological structures increase in size and complexity, and the drive to integrate them inevitably swells, it is clear that the level of consistency required for such navigation will become correspondingly difficult to maintain. While descriptive semantic representations are certainly a necessary component to any adequate ontology-based system, so long as ontology engineers rely solely on semantic information, without a sound ontological theory informing their modeling decisions, this goal will surely remain out of reach. In this paper we describe how Language and Computing nv (L&C), along with The Institute for Formal Ontology and Medical Information Sciences (IFOMIS), are working towards developing and implementing just such a theory, combining the open software architecture of L&C's LinkSuite TM with the philosophical rigor of IFOMIS's Basic Formal Ontology. In this way we aim to move beyond the more or less simple controlled vocabularies that have dominated the industry to date.

The integration of information resources in the life sciences is one of the most challenging ,problems ,facing bioinformatics ,today. We describe ,how Language and Computing nv, originally a developer of ontology-based natural language understanding systems for the healthcare domain, is developing a framework,for the integration of structured,data with unstructured information contained,in natural ,language ,texts. L&C’s LinkSuite™ ,combines ,the flexibility of a ,modular ,software ,architecture with ,an ontology ,based ,on rigorous philosophical and logical principles that is designed to comprehend,the basic formal ,relationships that structure both ,reality and the ways ,humans perceive and communicate,about reality.

this paper we describe how this standardization has already led to an improvement in the LinKBase structure that allows for a greater degree of internal coherence than ever before possible. We then show the use of this philosophical standardization for the purpose of mapping external databases to one another, using LinKBase as translation hub, with a greater degree of success than possible hitherto. We demonstrate how this offers a genuine advance over other application ontologies that have not submitted themselves to the demands of philosophical scrutiny