METHOD FOR DETECTING CRITICAL TRENDS AT MULTIPARAMETER CONTROL FOR A PATIENT AND CLINICAL DATA USING CLUSTERING

摘要

1. The device (10) analysis of physiological data to determine the status of an individual patient, containing:! an input unit (12) providing reception of many different physiological parameters of an individual patient; ! a classification unit (20) for comparing a plurality of physiological parameters with a multidimensional space having a plurality of regions corresponding to two or more states and determining a state of an individual patient in the region of physiological parameters mapped therein; and! an output unit (24) providing state transfer to the device user (10). ! 2. The analysis device (10) according to claim 1, characterized in that the classification unit (20) is configured to compare two or more sets of physiological parameters obtained at different time intervals and predict the future state of an individual patient based on the trend (trends ) extracted from the mappings. ! 3. The analysis device (10) according to claim 2, characterized in that the classification unit (20) is configured to perform time series analysis to determine the trend. ! 4. The analysis device (10) according to claim 2, characterized in that the classification unit (20) is configured to generate a trend by connecting two or more comparisons using a vector and extrapolating the subsequent comparison. ! 5. The analysis device (10) according to claim 2, characterized in that the physiological parameters associated with the multidimensional space contain one or more parameters from the group consisting of temperature; heart rate respiratory rate; systolic blood pressure; and white blood cell count. ! 6. The analysis device (10) according to claim 1, characterized