Ph. D

Stanford Medicine, Stanford, California, United States

Are you Ph. D?

Claim your profile

Publications (28)3.57 Total impact

  • [show abstract] [hide abstract]
    ABSTRACT: Objectives: Many groups are developing computer-interpretable clinical guidelines (CIGs) for use during clinical encounters. CIGs use "Task-Network Models" for representation but differ in their approaches to addressing particular modeling challenges. We have studied similarities and differences between CIGs in order to identify issues that must be resolved before a consensus on a set of common components can be developed.
    02/2004;
  • [show abstract] [hide abstract]
    ABSTRACT: this article we examine the formation of ties between biomedical imaging systems and clinical information systems in light of the cited necessity to enhance information sharing. Radiology can benefit from attracting individuals with a medical informatics background, just as informatics can benefit by attracting individuals with special expertise in imaging and image management. The potential for linkages between systems that manage images and those that deal with other patient information has been recognized since at least the early 1970s. For example, one early radiology system, developed by researchers at Massachusetts General Hospital, included functions for patient-examination scheduling and for film-library management. The system later evolved to include on-line reporting capabilities and links to a surgical-pathology accessioning system for automated access to confirmed pathologic diagnosis [2]. Developers of early radiology systems encountered several tasks that pertained to management of clinical data, independent of data type. Research issues included design of user interfaces, development of methods that supported structured data entry while attempting to avoid overly constrained linguistic expression, and construction of semiautomated tools for indexing and retrieving information. The traditional importance of images to clinical fields as diverse as dermatology, pathology, gastroenterology, surgery, and radiology assured that image management was destined to play a central role in clinical information systems design and construction once the storage and communications technologies to support such data types had matured. Thus, in recent years, the use of computersupported image data has become widespread in a variety of hospital departments. Such data are also...
    08/2002;
  • [show abstract] [hide abstract]
    ABSTRACT: this paper, we present our approach to implementing the GLIF3 GuideLine Execution Engine (GLEE) that tries to balance the requirements of sharability, flexibility and maintainability in guideline implementation
    08/2002;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: This paper also discusses our studies of algorithm creation, presents our classification of errors that can be introduced during guideline development, and describes the use of tools to catch some of these errors
    04/2002;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: Introduction Disease surveillance---including surveillance for nascent epidemics that could reflect occult acts of bioterrorism---requires the continuous analysis, interpretation, and feedback of systematically collected data. Surveillance can support many activities such as planning and research, but the most important reason for conducting surveillance is to identify changes in population health status that are amenable to control by intervention. The changes, or aberrations, must be detected from data sources that often have a highly variable baseline. Yet, because of the urgency of detecting incipient epidemics, methods for disease surveillance that are distinguished by their practicality, uniformity, and rapidity are preferred to those that may be most accurate and most complete. Methods for disease surveillance generally have relied on traditional statistical models (see Stroup, 1994). Such approaches typically take as input disease reports from passive surveillance and generate
    06/2001;
  • [show abstract] [hide abstract]
    ABSTRACT: This paper presents a framework for designing a decision-support system that addresses these issues. The primary goal of this framework is to allow rapid deployment of complex decision-support systems in a resource-constrained environment with limited administrative support
    05/2001;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: We evaluated the use of an updated version of the Guideline Interchange Format(GLIF), GLIF3, in the translation of clinical guidelines into an electronically encoded form. The goal of GLIF3 is to enable the effective sharing of such guidelines among various clinical institutions and settings. The newer version of GLIF was developed in response to limitations in the expressiveness of the earlier version, GLIF2. We videotaped and analyzed the encoding of two clinical guidelines both into GLIF3 and into GLIF2 by two medical informaticians. GLIF3 was found to be more robust than GLIF2 for representing the content and logical structure of the clinical guidelines studied. Our formative evaluation demonstrated that GLIF3's intended improvements in expressiveness were achieved. KEYWORDS: Knowledge representation, clinical guidelines, evaluation, cognitive analysis, knowledge sharing 3
    04/2001;
  • [show abstract] [hide abstract]
    ABSTRACT: This paper discusses the approach that InterMed has used to design a data model and expression language for GLIF3
    04/2001;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: hat GLIF would benefit from the inclusion of GEM elements that provide links to portions of the narrative document providing background information on which various decision steps and actions are based; we therefore extended the GLIF model by adding GEM elements to it. We evaluated the resulting model informally by applying it to an Alzheimer's disease treatment guideline [3]. Methods: We mapped GEM elements to GLIF class definitions, by finding those that were equivalent, or adding new attributes to GLIF classes representing relevant GEM documentation descriptors, or in some cases adding new classes to the GLIF model. We used both the Together/J UML tool and the Protg knowledge-modeling tool [4] to represent the merged GLIF-GEM ontology. While the UML tool was used to output a report of the ontology, including class diagrams and documentation, Protg was used to encode the Alzheimer guideline by creating instances of the ontology classes. We also marked the same guid
    04/2001;
  • [show abstract] [hide abstract]
    ABSTRACT: The time dimension is very important when reasoning with clinical data. Unfortunately, the task of temporal reasoning is inherently computationally expensive. As the problems tackled by clinical decision support systems become more varied, increased demands will be placed on the temporal reasoning component, which may lead to slow response times. This paper addresses this problem. It describes a temporal reasoning system called RASTA that uses a distributed algorithm that enables it to deal with large data sets. The algorithm also supports a variety of configuration options, enabling RASTA to deal with a range of application requirements.
    02/2001;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: This paper describes an approach for representing uncertain temporal information and reasoning with it, and has shown that this approach can be integrated with a temporal abstraction mechanism. Integrating indeterminacy support with a temporal abstraction mechanism is a novel feature of our system. The resulting system provides a powerful tool for dealing with temporal indeterminacy in clinical databases
    02/2001;
  • [show abstract] [hide abstract]
    ABSTRACT: be mapped to those used by the institutional applications. On levels B and C, ontological knowledge of the attributes of medical concepts and the relationships among medical concepts that go beyond the representation of basic data fields and concept hierarchy, is highly desirable. With such knowledge, we can perform rigorous type checking, range checking, and semantic checking (e.g., whether an expression refers to meaningful characteristics of a concept). The support of these ontological needs for guideline modeling is separated into three layers. The first layer, Core GLIF, is part of the GLIF specification language. It defines a standard interface to medical data and concepts. Core GLIF ontology defines how medical data and concepts should be referenced by guidelines. It also defines the scope of data items and how the data items acquire their values. Core GLIF is part of the GLIF specification. The second layer, Reference Information Model (RIM), is essential for guideline execut
    02/2001;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: sentation of domain-level concepts. Subclasses of macro represent different types of domain-level concepts and contain attributes particular to that concept. Macros enable declarative specification of a procedural pattern in the domain concept. The pattern is then realized by a set of primitive GLIF steps. The definition of the macro and its subclasses include the information necessary to map the macros to the corresponding pattern of primitive GLIF steps. This is a scalable approach for adding domain-level concepts in GLIF. The mapping also enables sharing of guidelines that contain macros. Sites and software tools that recognize the macro subclass can take advantage of the domain-level representation for authoring, visualization, and execution. Sites and tools that do not support a particular type of macro can still share and comprehend guidelines using the primitive steps. To develop the definition of macros, we analyzed guidelines of different types to identify patterns of domain
    02/2001;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: The National Guideline Clearinghouse (NGC) and its guideline classification system are significant contributions to the study of clinical practice guidelines (CPGs) and their incorporation into routine clinical care. The NGC classification system is primarily designed to support guideline retrieval. We believe that a guideline classification system should also support identification of features that relate to incorporation of executable CPGs into computerbased applications for sharing and delivering guideline -based advice. We have developed a proposed expansion of the NGC guideline classification for this purpose. The axes of the proposed scheme have implications for designing formal models and structures for representing and authoring CPGs. This scheme also has implications for future research. Introduction Clinical practice guidelines (CPGs) promise to improve the efficiency and quality of health care and to reduce practice variability. Professional societies, health care organiza...
    Proceedings / AMIA ... Annual Symposium. AMIA Symposium 09/2000;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: ion Levels GLIF3 enables modeling of guidelines at three levels of abstraction: A. Conceptual level. Guidelines at this level are represented as flowcharts that can be used for browsing, through guideline viewing programs. However, these guidelines cannot be used for computation in providing decision support. B. Computable level. Guidelines at this level may be verified for logical consistency and completeness. Expression syntax, definitions of patient data items and clinical actions, and flow of the algorithm are specified at this level. C. Implementable level. At this level, guidelines are appropriate for incorporation into particular institutional information system environments. Thus, these guidelines may incorporate non-sharable elements. Figure 1 shows part of the conceptual specification of a guideline for management of stable angina. 13 3.2 Changes in the object model The object model for GLIF3 defines new constructs and further structures GLIF2 constructs. Representat...
    08/2000;
  • [show abstract] [hide abstract]
    ABSTRACT: This paper presents a radiological guidance method and robotic system designed to facilitate renal surgical procedures. Combines the guidance procedure normally employed by practicing surgeons with a simple and cost-effective needle injection device. The needle injector exhibits an extremely low radiological profile while providing actuated needle motion. The system can be easily utilized by urologic surgeons and can be readily adaptable to any operating room.
    12/1999;
  • [show abstract] [hide abstract]
    ABSTRACT: The recent development of compact surgical robots and instrumentation raise the need of a suitable mechanism for positioning and support in the proximity of the operative field. The accuracy of the surgical instrument and surgical procedure heavily relies on the ability of the supporting device to provide a sturdy base under the payload and dynamics of the instrument.
    12/1999;
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: This paper presents a robotic system for precise needle insertion under radiological guidance for surgical interventions and for delivery of therapy. It is extremely compact and is compatible with portable X-ray units and computer tomography scanners. The system presents a modular structure comprising a global positioning module, a miniature robotic module, and a radiolucent needle driver module. This system is the newest member of a growing family of modular surgical robots under development. The system may be operated stand-alone under joystick control making it readily adaptable to any operating room, or under full image guided computer control.
    12/1999;
  • [show abstract] [hide abstract]
    ABSTRACT: Introduction Depression is among the most common psychiatric disorders in old age. In the older adult, chronic illness, functional impairment and stressful life events are all associated with depression. The most common precipitant is felt to be a medical illness. Other common precipitants are losing one's spouse, physical illness, lack of social interaction, and poor socioeconomic condition. Depression is thought to affect between 6 % and 14 % of primary care patients, and about 10 to 40 % of medical inpatients (Kalyanam & Shamoian, 1990). Several reports estimate that the prevalence of depression is high in nursing home residents; 10.4% for affective disorder (Rovner, German, Broadhead, Morriss, Brant, Blaustein, and Falstein, 1990). About 14% of the nursing home residents are recognized by nursing home physicians, and 65% are recognized by nursing staff and family, to suffer from depression. (Rovner, et. al 1990) Atypical de
    05/1999;
  • Mark A. Musen, Ph. D
    [show abstract] [hide abstract]
    ABSTRACT: ng specific medical information. It is not simply because of the successes of health telematics, but rather because of the infusion of the Internet and the World-Wide Web into the culture at large, that the next generation of health-care professionals is certain to be computer literate. Those clinicians assuredly will have high expectations for the use of information technology in their workplace, and it will be the sweeping expansion of online information resources for society in general that will fuel their specific expectations for advances in health telematics. Use of the World-Wide Web provides an excellent metaphor by which to frame discussion of clinicians' interactions with clinical information systems. The exponential growth of the Web became possible only when companies such as Yahoo! began to categorize myriad Web sites, providing a useful index into the overwhelming amount of information available online. Automated text-based indexing of Web pages by "crawler" programs cer
    03/1999;