Using causal reasoning in gait analysis

David E. Hirsch, Sheldon R. Simon, Tom Bylander, Michael A. Weintraub, Peter Szolovits

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

This paper describes a series of experiments in which expert diagnostic systems were constructed to analyze human pathologic gait. The difference between successive systems is the recognition of the need for both causal reasoning about the process of gait and experiential, associational knowledge that can control causal reasoning. The performance of the first system (Dr. Gait-]), which relies exclusively on associational knowledge, is quite limited. The second system (Dr. Gait-2), because it is based on a qualitative causal model of gait, overcame many of the difficulties faced by the first system, but its ability to diagnose cases is limited by the complexity of causal reasoning. The third system (Quawds), which we are currently developing, is an experiment in integrating causal reasoning with associational knowledge so that robust conclusions can be produced efficiently.

Original languageEnglish
Pages (from-to)337-356
Number of pages20
JournalApplied Artificial Intelligence
Volume3
Issue number2-3
DOIs
StatePublished - 1989
Externally publishedYes

Fingerprint

Dive into the research topics of 'Using causal reasoning in gait analysis'. Together they form a unique fingerprint.

Cite this