Application of the EAGOL Model to the Interpretation of Clinical Data

Artificial Intelligence in Medicine: Interpreting Clinical Data

Volume

Papers from the 1994 AAAI Spring Symposium

Contents

We have undertaken to conduct an analysis of the "dense, high volume data set typical of a critical care environment," which covers "a 12- hour period of the ICU treatment of a patient with severe respiratory distress." The methodology being used is that encompassed in the EAGOL system, which has been influenced, in part, by the model described by James V. Snyder, M.D. and Michael R. Pinsky, M.D. in their treatise on "Oxygen Transport in the Critically II1." We have been looking for an opportunity to redirect our attention from the complex, engineered systems that have been the primary focus of our work in recent years, to the medical arena, and are eager to investigate the appropriateness of the new EAGOL technology to the interpretation of ICU data. In the following paragraphs, we offer some highlights of the EAGOL model as developed and applied in other domains. Comments specific to the ICU problem with be given in italics.