Detection of artifacts in monitored trends in intensive care

In intensive care, decision-making is often based on trend analysis of phys iological parameters. Artifact detection is a pre-requisite for interpretat ion of trends both for clinical and research purposes. In this study, we de veloped and tested three methods of artifact detection in physiological dat a (systolic, mean and diastolic artery and pulmonary artery pressures, cent ral venous pressure, and peripheral temperature) using pre-filtered physiol ogical signals (2-min median filtering) from 41 patients after cardiac surg ery. These methods were: (1) the Rosner statistic; (2) slope detection with rules; and (3) comparison with a running median (median detection). After tuning the methods using data from 20 randomly chosen patients, the methods were tested using the data from the remaining patients. The results were c ompared with those obtained by manual identification of artifacts by three senior intensive care unit physicians. Out of an average of 22 480 data poi nts for each variable, the three observers labelled 0.98% (220 data points) as artifacts. The inter-observer agreement was good. The average (range) s ensitivity for artifact detection in all variables in the test database was 66% (33-92%) for the Rosner statistic, 64% (24-98%) for slope detection an d 72% (41-98%) for median detection. All methods had a high specificity (gr eater than or equal to 94%). Slope detection had the highest mean positive prediction rate (53%; 21-85%). When the performance was measured by the cos t function, slope detection and running median performed equally well and w ere superior to Rosner statistics for systemic arterial and central venous pressure and peripheral temperature. None of the methods produced acceptabl e results for pulmonary artery pressures. We conclude that median filtering of physiological variables is effective in removing artifacts. In post-ope rative cardiac surgery patients, the remaining artifacts are difficult to d etect among physiological and pathophysiological changes. This makes large databases for tuning artifact algorithms mandatory. Despite these limitatio ns, the performance of running median and slope detection were good in sele cted physiological variables. (C) 2000 Elsevier Science Ireland Ltd. All ri ghts reserved.