Webb, L. | bioRxiv Science

bioRxiv2021

Automated detection of artefacts in neonatal EEG with residual neural networks

Background and ObjectiveTo develop a computational algorithm that detects and identifies different artefact types in neonatal electroencephalography (EEG) signals. MethodsAs part of a larger algorithm, we trained a Residual Deep Neural Network on expert human annotations of EEG recordings from 79 term infants recorded in a neonatal intensive care unit (112 h of 18-channel recording). The network was trained using 10 fold cross validation in Matlab. Artefact types included: device interference, EMG, movement, electrode pop, and non-cortical biological rhythms. Performance was assessed by prediction statistics and further validated on a separate independent dataset of 13 term infants (143 h of 3-channel recording). EEG pre-processing steps, and other post-processing steps such as averaging probability over a temporal window, were also included in the algorithm. ResultsThe Residual Deep Neural Network showed high accuracy (95%) when distinguishing periods of clean, artefact-free EEG from any kind of artefact, with a median accuracy for individual patient of 91% (IQR: 81%-96%). The accuracy in identifying the five different types of artefacts ranged from 57%-92%, with electrode pop being the hardest to detect and EMG being the easiest. This reflected the proportion of artefact available in the training dataset. Misclassification as clean was low for each artefact type, ranging from 1%-11%. The detection accuracy was lower on the validation set (87%). We used the algorithm to show that EEG channels located near the vertex were the least susceptible to artefact. ConclusionArtefacts can be accurately and reliably identified in the neonatal EEG using a deep learning algorithm. Artefact detection algorithms can provide continuous bedside quality assessment and support EEG review by clinicians or analysis algorithms. HighlightsO_LIWe applied a Residual Deep Neural Network as part of an artefact detection algorithm in neonatal electroencephalograms. C_LIO_LIThe algorithm shows high accuracy in identifying artefactual data in general and for specific artefact types. C_LIO_LIEEG channels near the top of the head are less prone to artefact. C_LI

neuroscience↗

bioRxiv2020

Th2-like T-follicular helper cells promote functional antibody production during Plasmodium falciparum infection

The most advanced malaria vaccine only has approximately 30% efficacy in target populations, and avenues to improve next generation vaccines need to be identified. Functional antibodies are key effectors of both vaccine induced and naturally acquired immunity, with induction driven by T-follicular helper cells (TfH) CD4+ T cells. We assessed circulating TfH (cTfH) responses and functional antibody production in human volunteers experimentally infected with Plasmodium falciparum. Longitudinal single-cell RNA-sequencing of cTfH revealed peak transcriptional activation and clonal expansion of major cTfH subsets occurred at day 8 following infection and a population structure of cTfH capturing phenotypical subsets of Th1- and Th2-like cells. Among 40 volunteers, infection resulted in the emergence of activated ICOS+ cTfH cells. During peak infection, activation was restricted to Th2-like cTfH cells, while Th1-like cTfH cell activation occurred one week after treatment. To link cTfH activation to antibody induction, we assessed the magnitude and function of anti-malarial IgM and IgG after infection. The functional breadth and magnitude of parasite-specific antibodies was positively associated with Th2-cTfH activation. In contrast, Th1-cTfH activation was associated with the induction of plasma cells, which we have previously shown have a detrimental role in germinal cell formation and antibody development. Thus, we identified that during P. falciparum malaria infection in humans, the activation of Th2-cTfH but not other subsets correlates with the development of functional antibodies required for protective immunity. Data for the first time identify a specific cellular response that can be targeted by future malaria vaccines to improve antibody induction.

immunology↗