Clinical Decision Support with IM-Agents and ERMA Multi-agents

  • Authors:
  • Susan L. Mabry;Caleb R. Hug;Russell C. Roundy

  • Affiliations:
  • Whitworth College, Spokane, Washington;Whitworth College, Spokane, Washington;Sacred Heart Medical Center, Spokane, Washington

  • Venue:
  • CBMS '04 Proceedings of the 17th IEEE Symposium on Computer-Based Medical Systems
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

IM-Agents (Intelligent Monitoring Agents) provides a system infrastructure andproof of concept applications demonstrating meaningful diagnosis and interventionadvice for healthcare personnel. The problem domain is extremely complex, withuncertainties and interdependencies. The monitored environment for a patient iscomprised of a number of influences from numerous sources involving asynchronous,dynamic changes. Agents act as a collaborative team of specialists. Data isdynamically collected, filtered and integrated. Dynamic inference is accomplishedwithin agents through a hybrid approach of Fuzzy Logic, Causal Bayesian networks,trend analysis and qualitative logic. Emergency Medical Assistant (ERMA) componentsdemonstrate the system for the trauma environment with particular emphasis on typesof shock and stabilization of arterial blood gases. This approach for clinical decisionsupport systems shows promise for a number of critical care scenarios.