A Knowledge-Based Method for Temporal Abstraction of Clinical Data a Dissertation Submitted to the Program in Medical Informatio

A Knowledge-Based Method for Temporal Abstraction of Clinical Data a Dissertation Submitted to the Program in Medical Informatio

A KNOWLEDGE-BASED METHOD FOR TEMPORAL ABSTRACTION OF CLINICAL DATA A DISSERTATION SUBMITTED TO THE PROGRAM IN MEDICAL INFORMATION SCIENCES AND THE COMMITTEE ON GRADUATE STUDIES OF STANFORD UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY By Yuval Shahar October 1994 © Copyright by Yuval Shahar 1994 All Rights Reserved ii I certify that I have read this dissertation and that in my opinion it is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ___________________________________ Mark A. Musen (Principal Adviser) (Departments of Medicine and Computer Science) I certify that I have read this dissertation and that in my opinion it is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ___________________________________ Richard E. Fikes (Department of Computer Science) I certify that I have read this dissertation and that in my opinion it is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ___________________________________ Barbara Hayes-Roth (Department of Computer Science) Approved for the University Committee on Graduate Studies: ___________________________________ iii Abstract This dissertation describes a reasoning framework for knowledge-based systems, that is specific to the task of abstracting higher-level concepts from time-stamped data, but that is independent of any particular domain. I specify the theory underlying the framework by a logical model of time, parameters, events, and contexts: a knowledge-based temporal-abstraction theory. The domain-specific knowledge requirements and the semantics of the inference structure that I propose are well defined and can be instantiated for particular domains. I have applied my framework to the domain of clinical medicine. My goal is to create, from primary time-stamped patient data, interval-based temporal abstractions, such as "severe anemia for 3 weeks in the context of administering the drug AZT," and more complex patterns, involving several such intervals. These intervals can be used for planning interventions for diagnostic or therapeutic reasons, for monitoring plans during execution, and for creating high-level summaries of electronic medical records. Temporal abstractions are also helpful for explanation purposes. Finally, temporal abstractions can be a useful representation for comparing a therapy planner’s recommendation with that of the human user, when the goals in both plans can be described in terms of creation, maintenance, or avoidance of certain temporal patterns. I define a knowledge-based temporal-abstraction method that decomposes the task of abstracting higher-level, interval-based abstractions from input data into five subtasks. These subtasks are then solved by five separate, domain- independent, temporal-abstraction mechanisms. The temporal-abstraction mechanisms depend on four domain-specific knowledge types. The semantics of the four knowledge types and the role they play in each mechanism are defined formally. The knowledge needed to instantiate the temporal-abstraction mechanisms in any particular domain can be parameterized and can be acquired from domain experts manually or with automated tools. iv I present a computer program implementing the knowledge-based temporal- abstraction method: RÉSUMÉ. The architecture of the RÉSUMÉ system demonstrates several computational and organizational claims with respect to the desired use and representation of temporal-reasoning knowledge. The RÉSUMÉ system accepts input and returns output at all levels of abstraction; generates context-sensitive and controlled output; accepts and uses data out of temporal order, modifying a view of the past or of the present, as necessary; maintains several possible concurrent interpretations of the data; represents uncertainty in time and value; and facilitates its application to additional domains by editing only the domain-specific temporal-abstraction knowledge. The temporal-abstraction knowledge is organized in the RÉSUMÉ system as three ontologies (domain-specific theories of relations and properties) of parameters, events, and interpretation contexts, respectively, in each domain. I have evaluated the RÉSUMÉ system in the domains of protocol-based care, monitoring of children’s growth, and therapy of insulin-dependent diabetic patients. I have demonstrated that the knowledge required for instantiating the temporal-abstraction mechanisms can be acquired in a reasonable amount of time from domain experts, can be easily maintained, and can be used for creating application systems that solve the temporal-abstraction task in these domains. Understanding the knowledge required for abstracting clinical data over time is a useful undertaking. A clear specification of that knowledge, and its representation in an ontology specific to the task of abstracting concepts over time, as was done in the architecture of the RÉSUMÉ system, supports designing new medical and other knowledge-based systems that perform temporal- reasoning tasks. The formal specification of the temporal-abstraction knowledge also supports acquisition of that knowledge from domain experts, maintenance of that knowledge once acquired, reusing the problem-solving knowledge for temporal abstraction in other domains, and sharing the domain-specific knowledge with other problem solvers that might need access to the domain’s temporal-reasoning knowledge. v Acknowledgments My wife, Smadar, has been with me during the last three of my four graduate degrees, over two continents and three academic centers. That alone would be beyond most peoples’ endurance; yet, Smadar has always supported me in my quest for combining the disciplines of computer science and medicine, was amazingly patient, and has been a constant source of calmness. Our daughter, Lilach proved to be equally patient, and even our son, Tomer, who has joined our family only recently, has never been known to file a formal complaint. No acknowledgments would be complete without mentioning the help in continuing my graduate studies that I got, and the foundations in AI and general mathematics and computer science that I learned, from my former mentors: Larry Manevitz, Martin Golumbic, Larry Birnbaum and, especially, Jonathan Stavi. My main advisor, Mark Musen, has been supporting my ideas from the beginning. Mark helped me find my way around the rather new academic discipline of medical informatics, which turned out to be neither a subset of medicine nor an extension of computer science, two areas I was familiar with. Mark also was often my scientific editor, not an easy task. Barbara Hayes-Roth was highly encouraging from the beginning of this work, was always an excellent sounding board, and provided short but penetrating comments that influenced greatly the organization of my work. Richard Fikes’ meticulous examination of the details of my logical framework was immensely useful; it forced me to clarify my ideas so that the casual, though technical, reader can understand them unambiguously. Finally, a fourth, though unofficial, important reader and advisor of this thesis was Michael Kahn. Michael encouraged my interest in temporal reasoning in medicine, an area to which he himself contributed greatly, and agreed from the beginning of this research to support it with his advice; his comments were always sound and practical. I have also vi found in Michael’s Ph.D. thesis a model of clarity in writing and organization that I have tried, at least, to emulate. I have enjoyed greatly my stay at Stanford and its great environment—both its weather and its people. I feel indebted to Ted Shortliffe for arriving at that environment. I have been corresponding with Ted several years before I had arrived at Stanford, and it was due to his vision in creating the program in medical informatics and being the driving force behind it that I realized that my interests lay in this new interdisciplinary area. Ted also facilitated greatly my arrival at Stanford. Finally, when I was forming my thesis proposal, I found my discussions with Ted about his concepts as to how a proposal should look and the way it should be developed into a thesis as highly practical and useful. Working with everybody in the PROTÉGÉ-II/T-Helper gang was (and still is) a great experience. In particular, I would like to thank Samson Tu for all the detailed conversations we had on planning and temporal reasoning in medicine, which helped me greatly in focusing and implementing my ideas. Henrik (“super hacker”) Eriksson was always a great help too. I had also spent many pleasant hours discussing decision-analysis and other issues with John Egar. Finally, I had many highly relevant technical discussions with Amar Das, who is another brotherly spirit in the area of temporal reasoning in medicine, and who helped me find many of the reference sources. My written work would surely be unreadable if not for Lyn Dupré’s industrious editing of my papers over the years and of my drafts of this thesis. My friend Lynne Hollander often helped in this task, and supplied general encouragement and many interesting discussions. With respect to support and encouragement, I feel indebted also to Betty Riccio, Darlene Vian, Pat Swift and the excellent technical support of the SSRG group. Most of my work had been supported by the T-Helper project, funded under Grant No. HS06330 from the Agency for Health Care Policy and Research. The computing

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    328 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us