Reference: Downs, S. M.; Walker, M. G.; & Blum, R. L. Automated Summarization of On-Line Medical Records. 1986.
Abstract: The ability to automatically create patient summaries of arbitrary and appropriate complexity would represent an important advance. Such summaries would be useful adjunct to a patient record for clinical decision making, real-time patient monitoring, or for surveillance of quality of care. We describe a design for a automated summarization program to label and summarize major events in time-oriented medical records of systemic lupus erythematosus patients. The user interface provides an interactive, graphic representation of the record with active regions selectable by the user to display data supporting conclusions or to magnify a region showing grater detail of the patient record. The program design uses a hypothetico-deductive algorithm. Disease states are evoked based on attributes with abnormal values, and the evoked disease frames are confirmed by matching their templates against the patient recored using a temporal querying symtax. This produces likelihood ratios that are used for Bayesian updating of the evoked disease states. A knowledge base of definitions of medical objects and their casual relations uderlies the program. At the time of this writing (November 1985), we have just begun implementation of the program in Interlisp on Xerox 1108 workstations, using the Knowledge Engineering Environment (KEE (TM)) . Results of the implementation will be presented at the meeting.
Notes: Journal Memo.