Temporal Event Recognition: An Application to Left Ventricular Performance

Abstract

The problem of recognition of time-varying events, particularly motion concepts, is discussed and a framework for the abstraction of such concepts is briefly presented. We have designed and implemented a knowledge-based expert system for motion understanding that incorporates several novel features. A frame-based representation, which includes exception handling via similarity links and the organizational primitives IS-A and PART-OF is used to construct a knowledge base of temporal concepts. This knowledge base drives the recognition process that integrates the paradigms of hypothesize-and-trust and competition and co-operation among conceptually adjacent hypotheses. The recognition stra tegy also includes a hypothesis rating scheme that is amenable to empirical performance analysis. This framework has been applied to the problem of human left ventricular performance evaluation from image sequences, and thus the knowledge base also contains much of the current understanding of left ventricular dynamics. This paper describes some aspects of this application in detail and presents a complete analytic of a typical image sequence, emphasizing comparisons between human and computer analysis. Additional justification of the adequacy of the framework has been obtained using empirical studies of the hypothesis rating scheme.