"Expertness" from structured text?: RECONSIDER: a diagnostic prompting program

  • Authors:
  • Mark S. Tuttle;David D. Sherertz;Marsden S. Blois;Stuart Nelson

  • Affiliations:
  • Univ. Cal. San Francisco, San Francisco, CA;Section on Medical Information Science, UCSF;Univ. Cal. San Francisco, San Francisco, CA;State Univ. of New York at Stony Brook, Stony Brook, NY

  • Venue:
  • ANLC '83 Proceedings of the first conference on Applied natural language processing
  • Year:
  • 1983

Quantified Score

Hi-index 0.00

Visualization

Abstract

RECONSIDER is an interactive diagnostic prompting program which uses simple information retrieval techniques to prompt a physician regarding possible diagnoses, given a list of positive patient findings. Its knowledge base consists of "structured text" definitions of 3262 diseases and a synonym dictionary Patient findings, and their synonyms, are matched against inverted files of terms from the disease descriptions, the number and selectivity of the patient findings matching terms in a given disease description determine that disease's "score", and the matched diseases are sorted on this score to form a preliminary differential diagnosis. Definitions of diseases can be referenced for viewing by name, or by their position in a differential. While its first formal evaluation is not yet complete, the performance of RECONSIDER continues to exceed the expectations of user and designer alike.