Several clinical terminologies now utilize description logic to model the logical definitions of concepts. Recent editions of the Systematized Nomenclature of Medicine (SNOMED) have been developed using the description logic Ontylog. A significant design criterion for SNOMED is to keep concept expressions simple enough to be broadly usable by clinicians, while maintaining faithful representation of concept meaning. Motivated by this criterion, "role grouping" has been developed as an extension to the description logic Ontylog. This paper describes the problems that motivated the creation of role grouping, outlines the semantics of role grouping, illustrates the benefits of this construct with examples from SNOMED Clinical Terms, and provides an algorithm for determining normal forms for expressions involving role groups.

1 Introduction K-Rep is an object-oriented knowledge representation system based on description logic Woods and Schmolze, 1992]. K-Rep provides a language to express descriptions (e.g., of medical concepts such as drugs, treatments, and diseases), inference mechanisms for reasoning about descriptions and their relationship to one another, and considerable supporting infrastructure. This paper presents an overview of K-Rep's system architecture, which has been motivated in part by the needs of several applications in the area of clinical information systems Mays et al., 1996]. It has become clear that accurate and comprehensive controlled medical terminologies are crucial for advanced clinical information systems, e.g., to achieve comparable data across the enterprise in support of outcomes analysis and inter-operation of ancillary systems, conceptual indexing of clinical advice, and knowledge-based entry of orders for drugs, procedures, etc. Organization and maintenance of such a terminology is a formidable task: there are at least 100,000 concepts of initial interest , both national (or international) standardization and local customization of content must coexist, and the terminology will evolve signiicantly over time. The state of the art in medical terminologies is characterized by ad hoc deenitions, manual classiication, little inferen-tial power, and no provision for collaborative development. Therefore, the area of clinical information systems has become one of our key applications. In October 1994, after several years of experience with a Common Lisp implementation of the K-Rep description logic system Mays et al., 1991], we embarked upon a substantial redesign eeort to accomplish the following objectives: Scalability: K-Rep must support extremely large knowledge bases, potentially beyond the limits of virtual storage. ist independent of process termination or system shutdown. Distribution: Knowledge bases must be simultaneously accessible from multiple client systems in a networked environment. Embeddability: K-Rep must ooer terminology services as a seamlessly integrated component of larger application systems. Performance: K-Rep must be able to respond at interactive speeds in critical production environments. Authoring support: K-Rep must provide a professional development environment to facilitate the collaborative knowledge engineering and visualiza-tion activities of domain experts in an intuitive and appealing manner. Consistency maintenance: Revision and reclassii-cation of descriptions is essential, both for interactive knowledge base editing and for run-time infer-encing. Scope: K-Rep must accommodate a wide variety of information within descriptions, including both semantic roles and primitives, and non-semantic information such as bitmaps, synonyms, abbreviations , and codes. These considerations led us naturally to our current design point, illustrated in …

The K-Rep system, based on description logic, is used to represent and reason with large and expressive controlled medical terminologies. Expressive concept descriptions incorporate semantically precise definitions composed using logical operators, together with important non-semantic information such as synonyms and codes. Examples are drawn from our experience with K-Rep in modeling the InterMed laboratory terminology and also developing a large clinical terminology now in production use at Kaiser-Permanente. System-level scalability of performance is achieved through an object-oriented database system which efficiently maps persistent memory to virtual memory. Equally important is conceptual scalability-the ability to support collaborative development, organization, and visualization of a substantial terminology as it evolves over time. K-Rep addresses this need by logically completing concept definitions and automatically classifying concepts in a taxonomy via subsumption inferences. The K-Rep system includes a general-purpose GUI environment for terminology development and browsing, a custom interface for formulary term maintenance, a C+2 application program interface, and a distributed client-server mode which provides lightweight clients with efficient run-time access to K-Rep by means of a scripting language.

The problem of finding a set of assumptions which explain a given proposition is in general NP-hard, even when the background theory is an acyclic Horn theory. In this paper it is shown that when the background theory is acyclic Horn and its pseudo-completion ...

In this paper, we propose a new approach to intensional semantics of term subsumption languages. We introduce concept algebras, whose signatures are given by sets of primitive concepts, roles, and the operations of the language. For a given set of variables, standard results give us free algebras. We next define, for a given set of concept definitions, a term algebra, as the quotient of the free algebra by a congruence generated by the definitions. The ordering on this algebra is called descriptive subsumption (⊒Δ). We also construct a universal concept algebra, as a non-well-founded set given by the greatest fixed point of a certain equation. The ordering on this algebra is called structural subsumption (≽Δ). We prove there are unique mappings from the free algebras, to each of these, and establish that our method for classifying cycles in a term subsumption language, KREP, consists of constructing accessible pointed graphs, representing terms in the universal concept algebra, and checking a simulation relation between terms.

Some mistakes in spelling and typing produce correct words, such as typing "fig" when "fog" was intended. These errors are undetectable by traditional spelling correction techniques. In this paper we present a statistical technique capable of detecting and correcting some of these errors when they occur in sentences. Experimental results show that this technique is capable of detecting 76% of simple spelling errors and correcting 73%.

We examine the effect of increasing word list size on the error rate of spelling correctors. An experiment on a large body of text shows that an increase in the word list size decreases the error rate.

An object-centered knowledge representation language is presented, with examples of its use in the FAME (financial marketing expertise) system. This representation, based on the tradition of frames and semantic nets, provides automatic reclassification, value caching, multiple knowledge bases, and an interface to relational databases. The influence that the representation has had on the development of the FAME system is discussed

The integrated use of rule-based inference and an object centered knowledge representation (inheritance network) in a financial marketing consultation system is described. The rules provide a highly flexible pattern match capability and inference cycle for control. The inheritance network provides a convenient way to represent the conceptual structure of the domain. By merging the two techniques, the financial computation can be shared at the most general level, and rule inference is carried out at any appropriate level of generalization. Since domain knowledge is represented independently from control knowledge, knowledge about a particular problem-solving technique is decoupled from the conditions for its invocation. A large financial marketing system has been built, and examples are given.<>

F inancial marketing represents a new and challenging domain for expert system applications. The term financial marketing denotes financial decision processes used in marketing products and services of such a scale as to significantly impact a company's financial status. Customers interested in major computing technology acquisitions usually want financing plans to be safe, sound, and attractive. When making large equipment proposals, financial considerations often become as important as computing considerations. Typically, no "best" answer exists when developing appropriate acquisition planst that satisfy customer capacity requirements while meeting applicable financial objectives. The question is not simply to optimize some financial objective. Of necessity, any proposed solution will be the result of balancing competing goals. Furthermore, it would be insufficient to merely propose a plan; rather, we need explanations persuading decision makers that our proposals are reasonable-convincing financial explanations that "sell" our solution (specifically, an acquisition and financing plan). Consider the following simple example: A customer currently has an IBM 3081 KX3 processor installed, and is close to exceeding that machine's capacity. He can select a larger processor from the same product family (such as an IBM 3084 QX6), or one from a product family representing newer technology (such as an IBM 3090 200). The 3081 KX3 is still on lease, so lease termination penalties apply. The leasing firm offers a no-penalty lease termination if the 3081 KX3 is replaced by a _3084 QX6 leased from the same firm. On the < other hand, the 3090 200 has higher residual =8 value plus other financial incentives. In such a scenario, it is important to know who will make the acquisition decision and whether that person is measured on budget, earnings per share, or some other financial criterion. An earlier article described a rule-based approach to develop a financial marketing system.' While that effort demonstrated the feasibility of a knowledgebased approach, it proved inadequate as a vehicle for "scaling up" to a production system providing all the capabilities required. Financial marketing domain knowledge is considerable, but structured. Representthat initial version required many rules, but hid much

Of 28 patients with severe asthma routinely examined with upper gastrointestinal (UGI) x-ray films, 18 (64%) were found to have hiatus hernia and 13 (46%) were found to have demonstrable gastroesophageal reflux. These prevalences differed significantly (P less than .001) from those seen in a control population (19% and 5%, respectively). These data suggest that aspiration of gastric acid is a frequent incitant to severe asthma and that it should be routinely sought in the treatment-resistant asthmatic patient. Intensive medical regimens directed against reflux and acidity may bring notable improvement in asthma symptoms. Surgical restoration of effective lower esophageal spincter function has proved to be curative in other reported studies.

Melioidosis was diagnosed in association with lung cancer in a World War 2 veteran 26 years after geographic exposure. The case history illustrates the prolonged latency, difficult diagnosis, and resistance to therapy of chronic melioidosis. Implications are that chronic forms of the disease will continue to surface in veterans of Southeast Asian conflicts for decades.

A simple diagram for arterial blood gas tensions was constructed from values obtained from normal volunteers at rest, during voluntary breath-holding and during maximal voluntary hyperventilation. Thus, an appropriate arterial oxygen tension (PaO2) was established for any physiologic level of alveolar ventilation, as reflected by the arterial carbon dioxide tension (PaCO2). Blood gas tensions of 177 ambulatory patients with differing clinical types of pulmonary insufficiency were plotted against the diagram and the magnitude of the observed deviations from the normal range were compared to selected pulmonary function test results. The vast majority of test results exhibited lower oxygen tensions than predicted from the observed PaCO2. This actual-predicted PaO2 difference was directly related to the severity of airways obstruction in chronic obstructive pulmonary disease and to the severity of restriction in interstitial lung diseases. Patients in the early stages of lung disease had resting hypocarbia before the onset of hypoxemia or of significant pulmonary function test abnormality. The diagram provides a simple, rapid clinical method of diagnosing and estimating the severity of the blood gas disturbances which reflect V ˙ / Q ˙ mismatching at any physiologic level of alveolar ventilation. A simple diagram for arterial blood gas tensions was constructed from values obtained from normal volunteers at rest, during voluntary breath-holding and during maximal voluntary hyperventilation. Thus, an appropriate arterial oxygen tension (PaO2) was established for any physiologic level of alveolar ventilation, as reflected by the arterial carbon dioxide tension (PaCO2). Blood gas tensions of 177 ambulatory patients with differing clinical types of pulmonary insufficiency were plotted against the diagram and the magnitude of the observed deviations from the normal range were compared to selected pulmonary function test results. The vast majority of test results exhibited lower oxygen tensions than predicted from the observed PaCO2. This actual-predicted PaO2 difference was directly related to the severity of airways obstruction in chronic obstructive pulmonary disease and to the severity of restriction in interstitial lung diseases. Patients in the early stages of lung disease had resting hypocarbia before the onset of hypoxemia or of significant pulmonary function test abnormality. The diagram provides a simple, rapid clinical method of diagnosing and estimating the severity of the blood gas disturbances which reflect V ˙ / Q ˙ mismatching at any physiologic level of alveolar ventilation. The Role of Spontaneous Diastolic Depolarization in Second Degree A-V Block: The Mechanism of "Paroxysmal" A-V Block and of a New Form of Pseudo-supernormal A-V ConductionCHESTVol. 63Issue 5PreviewDr. Mauricio Rosenbaum and his coworkers describe (see page 678 ) the mechanism of an unusual type of A-V block which, in most instances, represents a special form of bradycardia-dependent bundle branch block. The failure of transmission of the atrial impulse to the ventricle is ascribed to hypopolarization and spontaneous depolarization in an injured region of one bundle branch with concomitant complete conduction block in the contralateral bundle branch. Full-Text PDF