EFFECTIVE DATA VALIDATION OF HIGH-FREQUENCY DATA - TIME-POINT-BASED, TIME-INTERVAL-BASED, AND TREND-BASED METHODS

Real-time systems for monitoring and therapy planning, which receive t heir data from on-line monitoring equipment and computer-based patient records, require reliable data. Data validation has to utilize and co mbine a set of fast methods to detect, eliminate, and repair faulty da ta, which may lead to life-threatening conclusions. The strength of da ta validation results from the combination of numerical and knowledge- based methods applied to both continuously-assessed high-frequency dat a and discontinuously-assessed data. Dealing with high-frequency data, examining single measurements is not sufficient. It is essential to t ake into account the behavior of parameters over time. We present time -point-, time-interval-, and trend-based methods for validation and re pair. These are complemented by time-independent methods for determini ng an overall reliability of measurements. The data validation benefit s from the temporal data-abstraction process, which provides automatic ally derived qualitative values and patterns. The temporal abstraction is oriented on a context-sensitive and expectation-guided principle. Additional knowledge derived from domain experts forms an essential pa rt for all of these methods. The methods are applied in the field of a rtificial ventilation of newborn infants. Examples from the real-time monitoring and therapy-planning system VIE-VENT illustrate the usefuln ess and effectiveness of the methods. (C) 1997 Elsevier Science Ltd.