Forehead electrodes sufficiently detect propofol-induced slow waves for the assessment of brain function after cardiac arrest

Abstract

In a recent study, we proposed a novel method to evaluate hypoxic ischemic encephalopathy (HIE) by assessing propofolinduced changes in the 19-channel electroencephalogram (EEG). The study suggested that patients with HIE are unable to generate EEG slow waves during propofol anesthesia 48 h after cardiac arrest (CA). Since a low number of electrodes would make the method clinically more practical, we now investigated whether our results received with a full EEG cap could be reproduced using only forehead electrodes. Experimental data from comatose post-CA patients (N = 10) were used. EEG was recorded approximately 48 h after CA using 19-channel EEG cap during a controlled propofol exposure. The slow wave activity was calculated separately for all electrodes and four forehead electrodes (Fp1, Fp2, F7, and F8) by determining the low-frequency (< 1 Hz) power of the EEG. HIE was defined by following the patients' recovery for six months. In patients without HIE (N = 6), propofol substantially increased (244 ± 91%, mean ± SD) the slow wave activity in forehead electrodes, whereas the patients with HIE (N = 4) were unable to produce such activity. The results received with forehead electrodes were similar to those of the full EEG cap. With the experimental pilot study data, the forehead electrodes were as capable as the full EEG cap in capturing the effect of HIE on propofol-induced slow wave activity. The finding offers potential in developing a clinically practical method for the early detection of HIE.

Early recovery of frontal EEG slow wave activity during propofol sedation predicts outcome after cardiac arrest

“…The findings are in line with results of a previous pilot study indicating similar association during controlled propofol sedation at 48 h after ROSC. 17,20 In the current study, we show the association in a larger patient group with routine propofol sedation and continuously recorded SWA from the ICU admission until 48 h from ROSC. We also utilize an approach clinically suitable for ICU environment including a practical EEG recording and automatic parametrization of SWA using C-Trend Index.…”

“…17 Further exploration of the data showed that the SWA was sufficiently captured by a reduced frontal electrode set. 20 In this prospective multicenter study, we investigated the accuracy of EEG SWA in predicting the outcome of comatose CA survivors during propofol sedation. We hypothesized that low SWA during propofol sedation in the first 48 h after CA would be associated with poor outcome.…”

A Correlation-Driven Mapping For Deep Learning application in detecting artifacts within the EEG

“…The proposed method requires no auxiliary reference signal and no specific feature extraction. The method was shown to be suitable in detecting artifacts from recordings collected with an easy-to-use electrode and device suitable for demanding clinical environment such as ICU where electrode attachment with minimum preparation is highly appreciated [25]. [26,27].…”

Automatic detection of artifacts in EEG by combining deep learning and histogram contour processing

“…At the same time, spatial correlation is also considered in this method, as it is applicable the recordings of multiple channels. Moreover, our aim was to train and validate a simple deep learning-based artifact removal method and explore its performance considering only forehead EEG data which has been proved to be effective in intensive care units (ICUs) where securely maintaining electrodes' contact and providing long-term highquality signal acquisition are essential [21].…”