Real-Time Individuation of Global Unsafe Anomalies and Alarm Activation

Abstract

In this paper we present the IGUANA (real time Individuation of Global Unsafe Anomalies and Alarm activation) framework which performs real-time analysis of clinical data to assess the instantaneous risk of a patient and identify dangerous situations. The proposed approach consists of two phases. As a first step, historical data is analyzed to build a model of both normal and unsafe situations, which can be tailored to specific behaviors of a given patient clinical situation. The model exploits a risk function to characterize the risk level of a patient by analyzing his/her vital signs. Then, an online classification phase is performed. A risk label is assigned to each measure by applying the most suitable model and an alarm is triggered for dangerous situations. To allow ubiquitous analysis, this step has been developed to run on mobile devices and its performance has been evaluated on both smart phone and personal computer. Experimental results, performed on 64 records of patients affected by different diseases, show the adaptability and the efficiency of the proposed approach.