Computer assessment of left ventricular wall motion: the ALVEN expert system

The current limited success of computer-assisted analysis of left ventricular (LV) dynamics is due to three main reasons: there is a strong tendency to remain within the realm of mathematical modeling for LV dynamics, and it is not at all clear that this is an adequate approach; in places where mathematical models alone may be insufficient, current computer science research into more sophisticated schemes is not yet complete, and thus, more basic research is required, particularly into artificial intelligence, representations of knowledge, and interpretation control structures, before applications such as LV performance can be solved, a view also stated in M. Boehm and K. Hoehne (in "Digital Image Processing in Medicine" (K. Hoehne, Ed.), Springer-Verlag, New York/Berlin, 1981); there is a distinct lack of knowledge about LV dynamics, in conjunction with disagreements about what is important to model and what terminology is to be used. Although each of these issues is addressed, the first two issues are concentrated on. Furthermore, a language for the expression of definitions for terminology has been designed, and a system for LV dynamics interpretation has been implemented.