Rough Set Methodology in Clinical Practice: Controlled Hospital Trial of the MET System

Abstract. Acute abdominal pain in childhood is a common but diagnostically challenging problem facing Emergency Department personnel. Experienced physicians use a combination of key clinical attributes to assess and triage a child, but residents and inexperienced physicians may lack this ability. In order to assist them, we used knowledge discovery techniques based on rough set theory to develop a clinical decision model to support the triage. The model was implemented as a module for the Mobile Emergency Triage system – a clinical decision support system for the rapid emergency triage of children with acute pain. The abdominal pain module underwent a clinical trial in the Emergency Department of a teaching hospital. The trial allowed us to compare in a prospective manner the accuracy of the system to the triage performance of the physicians and the residents. The preliminary results are very encouraging and they demonstrate validity of developing computer-based support tools for a c...

• Type: Conference
• Year: 2004
• Where: RSCTC
• Authors: Ken Farion, Wojtek Michalowski, Roman Slowinski, Szymon Wilk, Steven Rubin
• Keywords: Artificial Intelligence, Emergency Triage, Mobile Emergency Triage, RSCTC 2004, Triage