AUTHOR=Floyrac Aymeric , Doumergue Adrien , Legriel Stéphane , Deye Nicolas , Megarbane Bruno , Richard Alexandra , Meppiel Elodie , Masmoudi Sana , Lozeron Pierre , Vicaut Eric , Kubis Nathalie , Holcman David 

TITLE=Predicting neurological outcome after cardiac arrest by combining computational parameters extracted from standard and deviant responses from auditory evoked potentials

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 17 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2023.988394

DOI=10.3389/fnins.2023.988394

ISSN=1662-453X

ABSTRACT=Background: Severity of neuronal damage in comatose patients following anoxic brain injury is assessed through a multimodal evaluation. However, predicting the return to full consciousness of hospitalized post-anoxic comatose patients remains challenging. 
Methods: We present here a method based on machine learning to predict the return to consciousness and good neurological outcome based on the analysis of transient responses of the auditory evoked potentials. Data from event-related potentials (ERPs) were recorded non-invasively with four surface cranial electrodes at electro-encephalography (EEG), and treated retrospectively in a cohort of 29 post-cardiac arrest comatose patients, recorded between day 3 and day 6 following admission. We extracted several EEG features (standard deviation and similarity for standard auditory stimulations and number of extrema and oscillations for deviant auditory stimulations) from the time responses in a window of few hundreds of milliseconds. The responses to the standard and the deviant auditory stimulations were thus considered independently. By combining these features, we built a two-dimensional map to evaluate possible group clustering. Using Gaussian, K-neighbourhood and SVM classifiers, we could predict the neurological outcome of post-anoxic comatose patients, the validity of the method being tested by a cross-validation procedure. 
Results: Analysis in two-dimensions of the present data revealed two separated clusters of patients with good versus bad neurological outcome. When favouring the highest specificity of our mathematical algorithms (0.91), we found a sensitivity of 0.83 and an accuracy of 0.90, maintained when calculation was performed using data from only one central electrode. 
To conclude, statistics of standard and deviant responses considered separately provide complementary and confirmatory predictions of the outcome of anoxic comatose patients, better assessed when combining these features on a two-dimensional statistical map. The benefit of this method compared to classical EEG and ERP predictors should be tested in a large prospective cohort. If validated, this method could provide an alternative tool to intensivists, to better evaluate neurological outcome and improve patient management, without neurophysiologist assistance.