A knowledge-based approach to intelligent alarms in anesthesia.

The combination and aggregation of separate information sources to generate more ;intelligent' alarms using a knowledge-based approach is described. The approach gas developed as part of the knowledge-based anesthesia decision support system, AES-2, which monitors the patient's physiological state during anesthesia and eventually suggests therapeutic actions. Integration of the alarm system into a conventional information system is discussed. The state variable model used and the modeling of uncertainty are examined. The tools that support knowledge acquisition, design of knowledge-base prototypes. and test of the knowledge base of the AES-2 are described. A simulation example is given.