Automating ICU emergency signaling
Abstract
In this thesis, a recursive algorithm based on kernel mapping is applied to develop an automated, ICU Signaling. The method is portable and adaptive, and has lower complexity. Streams of different medical parameters are used to identify normal and abnormal conditions of individual patients in ICU. Using a system as such, the slightest of anomaly deviated from the norm, can be detected and alarmed, so that the medical team can take immediate and emergency action.