Skip to Main Content
Traditional medical decision support systems have not met with widespread use in clinical practice. Issues that have prevented clinical adoption include narrow domain focus, models which are not sufficiently deep, and lack of trust in the system's reliability. While the methodologies may be sound, the acquisition of sufficient medical domain knowledge is often the drawback. In this work, basic methodologies that have been applied in independent decision support systems are modified to function in a new domain in which computer-assisted decision making is focused on analysis of the electronic patient record with the goal of notifying the physician if significant changes have occurred in the state of the patient. The objective is not simply to report individual values that have fallen outside of normal ranges, but to examine the patient record using in-depth diagnostic models and provide three levels of alerts when significant changes occur.
Computer Science and Information Engineering, 2009 WRI World Congress on (Volume:5 )
Date of Conference: March 31 2009-April 2 2009