A fuzzy system for detecting distorted plethysmogram pulses in neonates and paediatric patients

Pulse oximetry is a useful, quick, non-invasive and widely used technology for monitoring oxygen saturation (SaO(2)) for neonates and paediatric patie nts. However, pulse oximetry is fraught with artefacts, causing false alarm s resulting from patient or probe movement. The shape of the plethysmogram is a useful visual indicator for determining the reliability of SaO(2) nume rical readings. If certain features could be defined that tag valid plethys mogram pulses, then automatic recognition of valid SaO(2) values can be att ained. We observed that the systolic upstroke time (t(1)), the diastolic ti me (t(2)) and heart rate (HR) extracted from the plethysmogram pulse consti tute features which can be used for detecting normal and distorted plethysm ogram pulses. Therefore, we developed a knowledge-based system using fuzzy logic for classifying plethysmogram pulses into two categories: valid and a rtefact. A total of 22497 pulse waveforms were used to define the system pa rameters. These were obtained from 13 patients with heart rates ranging bet ween 62 and 209 beats min(-1). A further 1420 waveforms obtained from anoth er four patients were used for testing the system, and visually classified into 833 (59%) valid and 587 (41%) distorted segments. The system was able to classify 679 (82%) valid segments and 543 (93%) distorted segments corre ctly. The calculations of the system's performance showed 82% sensitivity, 86% accuracy and 93% specificity. We, therefore, conclude that the algorith m used in this system can he implemented in its present form for real-time SaO(2) monitoring in intensive care for detecting valid and distorted pleth ysmogram pulses.