Resting State EEG-based biometrics for individual identification using convolutional neural networks

Abstract

Biometrics is a growing field, which permits identification of individuals by means of unique physical features. Electroencephalography (EEG)-based biometrics utilizes the small intra-personal differences and large inter-personal differences between individuals' brainwave patterns. In the past, such methods have used features derived from manually-designed procedures for this purpose. Another possibility is to use convolutional neural networks (CNN) to automatically extract an individual's best and most unique neural features and conduct classification, using EEG data derived from both Resting State with Open Eyes (REO) and Resting State with Closed Eyes (REC). Results indicate that this CNN-based joint-optimized EEG-based Biometric System yields a high degree of accuracy of identification (88%) for 10-class classification. Furthermore, rich inter-personal difference can be found using a very low frequency band (0-2Hz). Additionally, results suggest that the temporal portions over which subjects can be individualized is less than 200 ms.

Extraction of common task features in EEG-fMRI data using coupled tensor-tensor decomposition

“…The corresponding fMRI maps of these components are mostly showing the default mode network which include frontal gyrus, medial prefrontal cortex, amygdala, cerebral cortex, and areas of precuneus cortex and posterior cingulate cortex. The scalps maps of these alpha band activities resembles the resting EEG maps as shown in (Ma et al 2015) where also inter and intra subject differences were acknowledged in the resting According to the temporal signatures, most of the components had similar latencies to the canonical HRF (6 s) but there were two components which maintained consistent lags. Coupled component 5 in Fig.…”

Automated analysis of seizure semiology and brain electrical activity in presurgery evaluation of epilepsy: A focused survey

“…104 These architectures have been implemented using EEG signals for multiple applications such as biometrics, 121 EEG-based emotion Marker-based and marker-free systems Classification/prediction: Support vector machines, neural networks -Hand-crafted feature extraction and relies on assumptions -Controlled test situations -Using 2D videos fails to capture movements that are not included in the field of view of the camera; only processes seizures when the camera position is perpendicular to the patient's coronal plane -Algorithms required the presence of specific motions frequencies during seizures -Heavily constrained as they can only recognize seizures with significant movements and fail to capture subtle and fine-grained changes during motion seizures -Dependency on a sufficiently large amount of detected key points -Occlusion by objects, bed linens, or humans (family members and clinical staff) 104 These architectures have been implemented using EEG signals for multiple applications such as biometrics, 121 EEG-based emotion Marker-based and marker-free systems Classification/prediction: Support vector machines, neural networks -Hand-crafted feature extraction and relies on assumptions -Controlled test situations -Using 2D videos fails to capture movements that are not included in the field of view of the camera; only processes seizures when the camera position is perpendicular to the patient's coronal plane -Algorithms required the presence of specific motions frequencies during seizures -Heavily constrained as they can only recognize seizures with significant movements and fail to capture subtle and fine-grained changes during motion seizures -Dependency on a sufficiently large amount of detected key points -Occlusion by objects, bed linens, or humans (family members and clinical staff)…”

“…Deep learning has revolutionized computer vision through end-to-end learning, and applying it to time-series data is gaining increasing attention. 104 These architectures have been implemented using EEG signals for multiple applications such as biometrics, 121 EEG-based emotion recognition, 122 and modeling cognitive events. 123 These methodologies address the challenges of machine learning techniques based on temporal and frequency domains, wavelet transforms, or energy analysis and are robust to analyze high-dimensional data with a poor signal-to-noise ratio and considerable variability between individual subjects and recording sessions.…”

Adversarial Deep Learning in EEG Biometrics

“…Recent progress in EEG deep learning has capabilities to tackle this problem. However importantly, current deep learn- O ing models for EEG biometric identification are vastly evaluated by within-session (i.e., within-recording) cross-validation protocols [7][8][9][10]. Due to their deep and complex nature, these models are particularly prone to capturing recording-specific variability rather than individual neural biomarkers.…”