Graphical Access to Medical Expert Systems: II. Design of an Interface for Physicians^[*]

Summary

The ONCOCIN Interviewer program provides a graphical interface between physicians and an expert system that is designed to assist with therapy selection for patients receiving experimental cancer therapy. A principal goal has been to increase acceptance of advanced computer tools in a clinical setting. The interface has been developed for high-performance Lisp workstations and is tailored to the existing paper forms and practices of the outpatient clinic. To be flexible, the program makes use of a document formatting language to control a raster graphics display of medical forms, traditional paper versions of which have been used to track patient progress. The program utilizes a mouse input device coupled with a software-defined data entry approach that may be customized to the specific environment. The work described suggests ways in which high density graphics interfaces, with pointing devices rather than an emphasis on keyboards, may make decision support tools more useful to physicians and more acceptable to them.

Das ONCOCIN Interviewer-Programm bietet eine grafische Schnittstelle zwischen Ärzten und einem Expertensystem, die dazu dienen soll, Unterstützung bei der Auswahl einer Therapie für Patienten mit experimenteller Krebsbehandlung zu gewähren. Hauptziel ist die Erhöhung der Akzeptanz fortschrittlicher Computerwerkzeuge im Krankenhaus. Die Schnittstelle wurde für Hochleistungs-Lisp-Arbeitsstationen entwickelt und ist auf die in der Ambulanz gebräuchlichen Formulare und Praktiken zugeschnitten. Um flexibel zu sein, benutzt das Programm eine Dokumentenformatierungssprache zur Kontrolle rastergrafischer Darstellungen medizinischer Formulare, von denen herkömmlicherweise Papierversionen zur Überwachung des Krankheitsverlaufs benutzt wurden. Das Programm verwendet eine »Eingabemaus«, verbunden mit einem softwaregesteuerten Eingabeverfahren, das der spezifischen Umgebung angepaßt werden kann. Die vorliegende Arbeit schlägt Möglichkeiten vor, wie Schnittstellen mit hoher Dichte, bei denen Anzeigeeinrichtungen einer Tastatur vorgezogen werden, Werkzeuge zur Unterstützung der Entscheidungsfindung für Ärzte nützlicher und akzeptabler gestalten können.

Key-Words

Document Formatting - Software Input Devices - Lisp Workstations - Human-Computer Interaction - Expert Systems

Schlüssel-Wörter

Dokumentenformatierung - Software-Eingabevorrichtungen - Lisp-Arbeits-stationen - Mensch-Maschine-Wechselwirkung - Expertensysteme

^* This work has been supported by the National Library of Medicine under Grants LM-03395 and LM-00048 and by the Division of Research Resources under Grant RR-01613. Dr. Shortliffe is a Henry J. Kaiser Family Foundation Faculty Scholar in General Internal Medicine. Computer facilities were provided by the SUMEX-AIM resource under NIH Grant RR-007S5 and through an equipment gift from Xerox Corporation.

• REFERENCES

• 1 Bischoff M. B, Shortliffe E. H, Scott A. C, Carlson R. W, Jacobs C. D. Integration of a Computer-Based Consultant into the Clinical Setting. Proceed. 7th Ann. Symp. Computer Applic. Med. Care. New York: IEEE Press; 1983: 149-152.
• 2 Clancey W. J, Shortliffe E. H. (Eds) Reading in Medical Artificial Intelligence. Reading, MA: Addison-Wesley; 1984
• 3 Combs D. M, Musen M. A, Fagan L. M, Shortliffe E. H. Graphical Entry of Procedural and Inferential Knowledge. Proceed. AAMSI Congress 86, American Association for Medical Systems and Informatics, May. 1986
• 4 Gerring P. E, Shortliffe E. H, van Melle W. The Interviewer/Reasoner Model: An Approach to Improving System Responsiveness in Interactive AI Systems. Artif. Intell. Mag 1982; 3: 24-27.
• 5 Hickam D. H, Shortliffe E. H, Bisch-off M. B, Scott A. C, Jacobs C. D. The Treatment Advice of a Computer-based Cancer Chemotherapy Protocol Advisor. Ann. intern. Med 1985; 103: 928-936.
• 6 Lane C. D, Frisse M. E, Fagan L. M, Shortliffe E. H. Object-oriented Graphics in Medical Interface Design. Proceed. AAMSI Congress 86, American Association for Medical Systems and Informatics, May. 1986
• 7 Schultz J. R, Davis L. The Technology of PROMIS. Proceed. IEEE 1979; 67: 1237-1244.
• 8 Sheil B. Power Tools for Programmers. Datamation 29.No 1983; 2: 131-144.
• 9 Shortliffe E. H, Scott A. C, Bischoff M. B, Campbell A. C, van Melle W, Jacobs C. D. ONCOCIN: An Expert System for Oncology Protocol Management. Proceed. 7th Int. Joint Conf. Artif. Intell. (IJCAI)1981. 876-88l. Also Available in Buchanan B. G, Shortliffe E. H. Rule-Based Expert Systems: The MYCIN Experiments.. Reading, MA: Addison-Wesley; 1984
• 10 Teach R, Shortliffe E. H. An Analysis of Physician Attitudes Regarding Computer-Based Clinical Consultation Systems. Comput. biomed. Res 1981; 14: 542-558.
• 11 Teitelman W, Masinter L. The Interlisp Programming Environment. IEEE Computer 1981; 14: 25-34.
• 12 Tsuji S, Shortliffe E. H. Graphical Access to Medical Expert Systems: I Design of a Knowledge Engineer’s Interface. Meth. Inform. Med 1986; 25: 62-70.
• 13 van Melle W. A Domain-Independent System that Aids in Constructing Knowledge-Based Consultation Programs. PhD Diss. Stanford University; 1980