On beyond Gruber: "Ontologies" in today's biomedical information systems and the limits of OWL.

The word "ontology" was introduced to information systems when only closed-world reasoning systems were available. It was "borrowed" from philosophy, but literal links to its philosophical meaning were explicitly disavowed. Since then, open-world reasoning systems based on description logics have been developed, OWL has become a standard, and philosophical issues have been raised. The result has too often been confusion. The question "What statements are ontological" receives a variety of answers. A clearer vocabulary that is better suited to today's information systems is needed. The project to base ICD-11 on a "Common Ontology" required addressing this confusion. This paper sets out to systematise the lessons of that experience and subsequent discussions. We explore the semantics of open-world and closed-world systems. For specifying knowledge bases and software, we propose "invariants" or, more fully, "the first order invariant part of the background domain knowledge base" as an alternative to the words "ontology" and "ontological." We discuss the role and limitations of OWL and description logics and how they are complementary to closed world systems such as frames and to less formal "knowledge organisation systems". We illustrate why the conventions of classifications such as ICD cannot be formulated directly in OWL, but can be linked to OWL knowledge bases by queries. We contend that while OWL and description logics are major advances for representing invariants and terminologies, they must be combined with other technologies to represent broader background knowledge faithfully. The ICD-11 architecture is one approach. We argue that such hybrid architectures can and should be developed further.

Scrutinizing SNOMED CT's Ability to Reconcile Clinical Language Ambiguities with an Ontology Representation.

An important SNOMED CT use case is to support semantic interoperability between electronic health records and aggregation terminologies such as ICD. From the ongoing alignment exercise between SNOMED CT and the new version of ICD, now in its pre-final form, we studied whether the ambiguity of clinical language as displayed by SNOMED CT synonyms hampers the quality of SNOMED CT axioms following the SNOMED CT "concept model". We measure the rate of synonyms in the semantic misalignment between classes from the chapter on circulatory diseases of the ICD-11 beta version and SNOMED CT concepts with the same description names. Our study confirms that SNOMED CT synonyms are ambiguous and that there is a need to increase the number of SNOMED CT fully defined representations of Fully Specified Names (FSN), and of synonyms independently of their relations.

Interface Terminologies, Reference Terminologies and Aggregation Terminologies: A Strategy for Better Integration.

The time has come to end unproductive competitions among different types of biomedical terminology artefacts. Tools and strategies to create the foundation of a seamless environment covering clinical jargon, clinical terminologies, and classifications are necessary. Whereas language processing relies on human interface terminologies, which represent clinical jargon, their link to reference terminologies such as SNOMED CT is essential to guarantee semantic interoperability. There is also a need for interoperation between reference and aggregation terminologies. Simple mappings between nodes are not enough, because the three kinds of terminology systems represent different things: reference terminologies focus on context-free descriptions of classes of entities of a domain; aggregation terminologies contain rules that enforce the principle of single hierarchies and disjoint classes; interface terminologies represent the language used in a domain. We propose a model that aims at providing a better flow of standardized information, addressing multiple use cases in health care including clinical research, epidemiology, care management, and reimbursement.

Representing ICD-11 JLMMS Using IHTSDO Representation Formalisms.

It is investigated whether the content of the Joint Linearization for Mortality and Morbidity Statistics of the 11th ICD revision can be semantically represented by formalisms acting on the clinical terminology SNOMED CT, viz. the IHTSDO Compositional Grammar (CG) and the Expression Constraint Language (ECL). Whereas CG provides a composition syntax for building coordinated SNOMED CT expressions, ECL provides a powerful query mechanism. Both formalisms can be leveraged to guarantee inter-operation between an ontology-based terminology like SNOMED CT and a statistical classification like ICD, characterized by single hierarchies and exhaustive, mutually exclusive classes. We test the feasibility of the method on the circulatory chapter of ICD-11 JLMMS.

ICD-11 (JLMMS) and SCT Inter-Operation.

The goal of this work is to contribute to a smooth and semantically sound inter-operability between the ICD-11 (International Classification of Diseases-11th revision Joint Linearization for Mortality, Morbidity and Statistics) and SNOMED CT (SCT). To guarantee such inter-operation between a classification, characterized by a single hierarchy of mutually exclusive and exhaustive classes, as is the JLMMS successor of ICD-10 on the one hand, and the multi-hierarchical, ontology-based clinical terminology SCT on the other hand, we use ontology axioms that logically express generalizable truths. This is expressed by the compositional grammar of SCT, together with queries on axiomsof SCT. We test the feasibility of the method on the circulatory chapter of ICD-11 JLMMS and present limitations and results.

Semantic Alignment between ICD-11 and SNOMED CT.

Due to fundamental differences in design and editorial policies, semantic interoperability between two de facto standard terminologies in the healthcare domain--the International Classification of Diseases (ICD) and SNOMED CT (SCT), requires combining two different approaches: (i) axiom-based, which states logically what is universally true, using an ontology language such as OWL; (ii) rule-based, expressed as queries on the axiom-based knowledge. We present the ICD-SCT harmonization process including: a) a new architecture for ICD-11, b) a protocol for the semantic alignment of ICD and SCT, and c) preliminary results of the alignment applied to more than half the domain currently covered by the draft ICD-11.

Using the wisdom of the crowds to find critical errors in biomedical ontologies: a study of SNOMED CT.

OBJECTIVES: The verification of biomedical ontologies is an arduous process that typically involves peer review by subject-matter experts. This work evaluated the ability of crowdsourcing methods to detect errors in SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms) and to address the challenges of scalable ontology verification. METHODS: We developed a methodology to crowdsource ontology verification that uses micro-tasking combined with a Bayesian classifier. We then conducted a prospective study in which both the crowd and domain experts verified a subset of SNOMED CT comprising 200 taxonomic relationships. RESULTS: The crowd identified errors as well as any single expert at about one-quarter of the cost. The inter-rater agreement (κ) between the crowd and the experts was 0.58; the inter-rater agreement between experts themselves was 0.59, suggesting that the crowd is nearly indistinguishable from any one expert. Furthermore, the crowd identified 39 previously undiscovered, critical errors in SNOMED CT (eg, 'septic shock is a soft-tissue infection'). DISCUSSION: The results show that the crowd can indeed identify errors in SNOMED CT that experts also find, and the results suggest that our method will likely perform well on similar ontologies. The crowd may be particularly useful in situations where an expert is unavailable, budget is limited, or an ontology is too large for manual error checking. Finally, our results suggest that the online anonymous crowd could successfully complete other domain-specific tasks. CONCLUSIONS: We have demonstrated that the crowd can address the challenges of scalable ontology verification, completing not only intuitive, common-sense tasks, but also expert-level, knowledge-intensive tasks.

What's in a class? Lessons learnt from the ICD - SNOMED CT harmonisation.

The upcoming ICD-11 will be harmonized with SNOMED CT via a common ontological layer (CO). We provide evidence for our hypothesis that this cannot be appropriately done by simple ontology alignment, due to diverging ontological commitment between the two terminology systems. Whereas the common ontology describes clinical situations, ICD-11 linearization codes are best to be interpreted as diagnostic statements. For the binding between ICD codes and classes from the ontological layer, a query-based approach is favoured.

ICD-11 and SNOMED CT Common Ontology: circulatory system.

The improvement of semantic interoperability between data in electronic health records and aggregated data for health statistics requires efforts to carefully align the two domain terminologies ICD and SNOMED CT. Both represent a new generation of ontology-based terminologies and classifications. The proposed alignment of these two systems and, in consequence, the validity of their cross-utilisation, requires a specific resource, named Common Ontology. We present the ICD-11 SNOMED CT Common Ontology building process including: a) the principles proposed for aligning the two systems with the help of a common model of meaning, b) the design of this common ontology, and c) preliminary results of the application to the diseases of the circulatory system.

Sharing ontology between ICD 11 and SNOMED CT will enable seamless re-use and semantic interoperability.

In order to support semantic interoperability in eHealth systems, domain terminologies need to be carefully designed. SNOMED CT and the upcoming ICD-11 represent a new generation of ontology-based terminologies and classifications. The proposed alignment of these two systems and, in consequence, the validity of their cross-utilisation requires a thorough analysis of the intended meaning of their representational units. We present the ICD11 SNOMED CT harmonization process including: a) the clarification of the interpretation of codes in both systems as representing situations rather than conditions, b) the principles proposed for aligning the two systems with the help of a common ontology, c) the high level design of this common ontology, and d) further ontology-driven issues that have arisen in the course of this work.

Analysing Syntactic Regularities and Irregularities in SNOMED-CT.

MOTIVATION: In this paper we demonstrate the usage of RIO; a framework for detecting syntactic regularities using cluster analysis of the entities in the signature of an ontology. Quality assurance in ontologies is vital for their use in real applications, as well as a complex and difficult task. It is also important to have such methods and tools when the ontology lacks documentation and the user cannot consult the ontology developers to understand its construction. One aspect of quality assurance is checking how well an ontology complies with established 'coding standards'; is the ontology regular in how descriptions of different types of entities are axiomatised? Is there a similar way to describe them and are there any corner cases that are not covered by a pattern? Detection of regularities and irregularities in axiom patterns should provide ontology authors and quality inspectors with a level of abstraction such that compliance to coding standards can be automated. However, there is a lack of such reverse ontology engineering methods and tools. RESULTS: RIO framework allows regularities to be detected in an OWL ontology, i.e. repetitive structures in the axioms of an ontology. We describe the use of standard machine learning approaches to make clusters of similar entities and generalise over their axioms to find regularities. This abstraction allows matches to, and deviations from, an ontology's patterns to be shown. We demonstrate its usage with the inspection of three modules from SNOMED-CT, a large medical terminology, that cover "Present" and "Absent" findings, as well as "Chronic" and "Acute" findings. The module sizes are 5 065, 20 688 and 19 812 asserted axioms. They are analysed in terms of their types and number of regularities and irregularities in the asserted axioms of the ontology. The analysis showed that some modules of the terminology, which were expected to instantiate a pattern described in the SNOMED-CT technical guide, were found to have a high number of regularity deviations. A subset of these were categorised as "design defects" by verifying them with past work on the quality assurance of SNOMED-CT. These were mainly incomplete descriptions. In the worst case, the expected patterns described in the technical guide were followed by only 5% of the axioms in the module. CONCLUSION: It is possible to automatically detect regularities and then inspect irregularities in an ontology. We argue that RIO is a tool to find and report such matches and mismatches, for evaluations by the domain experts. We have demonstrated that standard clustering techniques from machine learning can offer a tool in the drive for quality assurance in ontologies. AVAILABILITY: http://riotool.sourceforge.net/ CONTACT: http://eleni.mikroyannidi@manchester.ac.uk, http://robert.stevens@manchehster.ac.uk.

Competing interpretations of disorder codes in SNOMED CT and ICD.

Under ontological scrutiny we have identified two competing interpretations of disorder concepts in SNOMED. Should codes be interpreted as representing pathological conditions themselves or the situations in which a patient has those conditions? This difference has significant implications for the proposed harmonization between SNOMED CT and the new ICD-11 disease classification and indeed for any systematic review of the correctness of the SNOMED CT hierarchies. Conditions themselves are distinct, whereas in any given situation a patient may have more than one condition. In such cases, SNOMED codes that represent combinations of conditions - which can be regarded as "additive" - are evidence for interpreting the codes as referring to situations. There are clearly some such codes. We conducted a survey to determine the extent of this phenomenon. Three criteria were used - analysis of the SNOMED CT fully specified name, the corresponding logical definition, and the children of the concept under scrutiny. All three showed that at least 11% of concepts met our criteria for representing situations rather than conditions, with a satisfactory inter-rater reliability for the first two. We, therefore, conclude that if a uniform interpretation of SNOMED disorder codes is desired, they should be interpreted as representing situations.

Lexically suggest, logically define: quality assurance of the use of qualifiers and expected results of post-coordination in SNOMED CT.

A study of the use of common qualifiers in SNOMED CT definitions and the resulting classification was undertaken using combined lexical and semantic techniques. The accuracy of SNOMED authors in formulating definitions for pre-coordinated concepts was taken as a proxy for the expected accuracy of users formulating post-coordinated expressions. The study focused on "acute" and "chronic" as used within a module based on the UMLS CORE Problem List and using the pattern of SNOMED CT's definition Acute disease and Chronic disease. Scripts were used to identify potential candidate concepts whose names suggested that they should be classified as acute or chronic findings. The potential candidates were filtered by local clinical experts to eliminate spurious lexical matches. Scripts were then use to determine which of the filtered candidates were not classified under acute or chronic findings as expected. The results were that 28% and 20% of candidate chronic and acute concepts, respectively, were not so classified. Of these candidate misclassifications, the large majority occurred because "acute" and "chronic" are sometimes specified by qualifiers for clinical course and sometimes for morphology, a fact mentioned but not fully detailed in the User Guide distributed with the SNOMED releases. This heterogeneous representation reflects a potential conflict between common usage in patient care and SNOMED's origins in pathology. Other incidental findings included questions about the qualifier hierarchies themselves and issues with the underlying model for anatomy. The effort required for the study was kept modest by using module extraction and scripts, showing that such quality assurance of SNOMED is practical. The results of a preliminary study using proxy measures must be taken with caution. However, the high rate of misclassification indicates that, until the specifications for qualifiers are better documented and/or brought more in line with common clinical usage, anyone attempting to use post-coordination in SNOMED CT must be aware that there are significant pitfalls.

Getting the foot out of the pelvis: modeling problems affecting use of SNOMED CT hierarchies in practical applications.

OBJECTIVES: (a) To determine the extent and range of errors and issues in the Systematised Nomenclature of Medicine-Clinical Terms (SNOMED CT) hierarchies as they affect two practical projects. (b) To determine the origin of issues raised and propose methods to address them. METHODS: The hierarchies for concepts in the Core Problem List Subset published by the Unified Medical Language System were examined for their appropriateness in two applications. Anomalies were traced to their source to determine whether they were simple local errors, systematic inferences propagated by SNOMED's classification process, or the result of problems with SNOMED's schemas. Conclusions were confirmed by showing that altering the root cause and reclassifying had the intended effects, and not others. MAIN RESULTS: Major problems were encountered, involving concepts central to medicine including myocardial infarction, diabetes, and hypertension. Most of the issues raised were systematic. Some exposed fundamental errors in SNOMED's schemas, particularly with regards to anatomy. In many cases, the root cause could only be identified and corrected with the aid of a classifier. LIMITATIONS: This is a preliminary 'experiment of opportunity.' The results are not exhaustive; nor is consensus on all points definitive. CONCLUSIONS: The SNOMED CT hierarchies cannot be relied upon in their present state in our applications. However, systematic quality assurance and correction are possible and practical but require sound techniques analogous to software engineering and combined lexical and semantic techniques. Until this is done, anyone using SNOMED codes should exercise caution. Errors in the hierarchies, or attempts to compensate for them, are likely to compromise interoperability and meaningful use.

Integration of tools for binding archetypes to SNOMED CT.

BACKGROUND: The Archetype formalism and the associated Archetype Definition Language have been proposed as an ISO standard for specifying models of components of electronic healthcare records as a means of achieving interoperability between clinical systems. This paper presents an archetype editor with support for manual or semi-automatic creation of bindings between archetypes and terminology systems. METHODS: Lexical and semantic methods are applied in order to obtain automatic mapping suggestions. Information visualisation methods are also used to assist the user in exploration and selection of mappings. RESULTS: An integrated tool for archetype authoring, semi-automatic SNOMED CT terminology binding assistance and terminology visualization was created and released as open source. CONCLUSION: Finding the right terms to bind is a difficult task but the effort to achieve terminology bindings may be reduced with the help of the described approach. The methods and tools presented are general, but here only bindings between SNOMED CT and archetypes based on the openEHR reference model are presented in detail.

Why do it the hard way? The case for an expressive description logic for SNOMED.

There has been major progress both in description logics and ontology design since SNOMED was originally developed. The emergence of the standard Web Ontology language in its latest revision, OWL 1.1 is leading to a rapid proliferation of tools. Combined with the increase in computing power in the past two decades, these developments mean that many of the restrictions that limited SNOMED's original formulation no longer need apply. We argue that many of the difficulties identified in SNOMED could be more easily dealt with using a more expressive language than that in which SNOMED was originally, and still is, formulated. The use of a more expressive language would bring major benefits including a uniform structure for context and negation. The result would be easier to use and would simplify developing software and formulating queries.

A deployment and Research Roadmap for Semantic Interoperability: the EU semantic health project.

The purpose of this EU funded project is to describe a short and medium term Research and Deployment Roadmap for Semantic Interoperability in e-health. It started by defining 4 levels and 3 dimensions for Semantic Interoperability. The vision is to reconcile the needs for the direct patient care safety, biomedical and clinical research and for public health by the reuse of direct care data: from gene to individuals and populations. The methodology is presented and preliminary results and milestones for the short and the long term are set. We conclude by statements on the main characteristics and needs of the roadmap to sustain better health for individual and populations in the changing EU health care systems.

Semantic issues in integrating data from different models to achieve data interoperability.

Matching clinical data to codes in controlled terminologies is the first step towards achieving standardisation of data for safe and accurate data interoperability. The MoST automated system was used to generate a list of candidate SNOMED CT code mappings. The paper discusses the semantic issues which arose when generating lexical and semantic matches of terms from the archetype model to relevant SNOMED codes. It also discusses some of the solutions that were developed to address the issues. The aim of the paper is to highlight the need to be flexible when integrating data from two separate models. However, the paper also stresses that the context and semantics of the data in either model should be taken into consideration at all times to increase the chances of true positives and reduce the occurrence of false negatives.

What's in a code? Towards a formal account of the relation of ontologies and coding systems.

Terminologies are increasingly based on "ontologies" developed in description logics and related languages such as the new Web Ontology Language, OWL. The use of description logic has been expected to reduce ambiguity and make it easier determine logical equivalence, deal with negation, and specify EHRs. However, this promise has not been fully realised: in part because early description logics were relatively inexpressive, in part, because the relation between coding systems, EHRs, and ontologies expressed in description logics has not been fully understood. This paper presents a unifying approach using the expressive formalisms available in the latest version of OWL, OWL 1.1.

Use of OWL-based tools to aid message development and maintenance.

The National Health Service National Programme for Information Technology is currently developing messaging standards to make available key information about an individual patient's care across and within different health-care settings. The complexity of this task makes it imperative that adequate tooling support is made available so as to aid consistency of design, implementation, conformance testing, and maintenance. Connecting for Health commissioned a proof of concept study to investigate whether the Object Web Language (OWL) and its associated tools could provide a basis for the message development and implementation environments. This study indicated that OWL is very much suited to such a task and should be seriously considered.