Deep Learning Techniques to Improve Intraoperative Awareness Detection from Electroencephalographic Signals

  title={Deep Learning Techniques to Improve Intraoperative Awareness Detection from Electroencephalographic Signals},
  author={Oleksii Avilov and S{\'e}bastien Rimbert and Anton Popov and Laurent Bougrain},
  journal={2020 42nd Annual International Conference of the IEEE Engineering in Medicine \& Biology Society (EMBC)},
  • O. Avilov, Sébastien Rimbert, +1 author L. Bougrain
  • Published 1 July 2020
  • Computer Science, Medicine
  • 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)
Every year, millions of patients regain conscious- ness during surgery and can potentially suffer from post-traumatic disorders. We recently showed that the detection of motor activity during a median nerve stimulation from electroencephalographic (EEG) signals could be used to alert the medical staff that a patient is waking up and trying to move under general anesthesia [1], [2]. In this work, we measure the accuracy and false positive rate in detecting motor imagery of several deep learning… Expand

Figures and Tables from this paper

Optimizing Winograd-Based Convolution with Tensor Cores
  • Junhong Liu, Dongxu Yang, Junjie Lai
  • Computer Science
  • ICPP
  • 2021
An optimized mixed precision F(6 × 6, 3 × 3) Winograd convolution implementation on NVIDIA Ampere GPUs using Tensor Cores and integrates into NVIDIA TensorRT, which is a C++ inference library on GPUs provided by NVIDIA, as custom layer plugins. Expand
Implementation of Artificial Neural Networks for the Nepta-Uranian Interplanetary (NUIP) Mission
A design of deep neural networks, namely convolutional neural networks (CNN) and recurrent neural Networks (RNN), capable of predicting optimal control actions and image classification during the Nepta-Uranan interplanetary mission are proposed. Expand


Innovative Brain-Computer Interface based on motor cortex activity to detect accidental awareness during general anesthesia
The goal of this forward-looking study is to investigate the motor activity behavior with step-wise increase of propofol doses in 4 healthy subjects and provide a proof of concept for such an innovative BCI. Expand
Decoding motor responses from the EEG during altered states of consciousness induced by propofol.
The results show the potential of using a BCI to detect intra-operative awareness and justify further development of this paradigm, and the relationship between motor responses and consciousness and its clinical relevance for intraoperative awareness requires further investigation. Expand
Deep learning for electroencephalogram (EEG) classification tasks: a review.
Practical suggestions on the selection of many hyperparameters are provided in the hope that they will promote or guide the deployment of deep learning to EEG datasets in future research. Expand
Median Nerve Stimulation Based BCI: A New Approach to Detect Intraoperative Awareness During General Anesthesia
The results show that MNS may provide a foundation for an innovative BCI that would allow the detection of AAGA, and investigate the specific modulations that M NS causes in the motor cortex and confirm that they can be altered by an intention of movement. Expand
EEGNet: A Compact Convolutional Neural Network for EEG-based Brain-Computer Interfaces
Objective: Brain computer interfaces (BCI) enable direct communication with a computer, using neural activity as the control signal. This neural signal is generally chosen from a variety ofExpand
Deep learning with convolutional neural networks for EEG decoding and visualization
This study shows how to design and train convolutional neural networks to decode task‐related information from the raw EEG without handcrafted features and highlights the potential of deep ConvNets combined with advanced visualization techniques for EEG‐based brain mapping. Expand
Common Spatial Pattern Method for Channel Selelction in Motor Imagery Based Brain-computer Interface
A method for channel reduction in Mi-based BCI is presented and a classification algorithm was developed by means of combining linear discriminant analysis towards event-related desynchronization (ERD) and readiness potential (RP). Expand
Functional connectivity changes with concentration of sevoflurane anesthesia
Low-frequency oscillations (<0.08 Hz) have been detected in functional magnetic resonance imaging studies, and appear to be synchronized between functionally related areas. The effect of anestheticExpand
Multiclass Brain–Computer Interface Classification by Riemannian Geometry
A new classification framework for brain-computer interface (BCI) based on motor imagery using spatial covariance matrices as EEG signal descriptors and relying on Riemannian geometry to directly classify these matrices using the topology of the manifold of symmetric and positive definite matrices. Expand
Disruption of cortical network activity by the general anaesthetic isoflurane
Disruption of horizontal activity spread and of layer 1 facilitation of thalamo-cortical responses likely contribute to the mechanism by which suppression of cortical feedback connections disrupts sensory awareness under anaesthesia. Expand