Electroencephalogram Based Medical Biometrics using Machine Learning: Assessment of Different Color Stimuli

Abstract

A methodology of medical signal-based biometrics has been proposed in this paper for implementing a human identification system controlled by electroencephalogram in respect of different color stimuli. The advantage of biosignal based biometrics is that they provide more efficient operation in simple experimental condition to ensure accurate identification. Red, Green, Blue (primary colors) and Yellow (secondary color) were chosen as the color stimuli for making more comfortable EEG regenerating environment. Four supervised classification models, namely, Logistic Regression (LR), K-Nearest Neighbor (KNN), Support Vector Machine (SVM) and Random Forest Classifier (RFC) were trained and tested for assessing the performance of the EEG based biometric identification, with fivefold cross-validation. Four different measures (sensitivity, specificity, accuracy and area under the receiver operating characteristic curve) were used to evaluate the overall performance. The results suggested that Blue color stimuli perform the best among all the color stimulus with an accuracy ranging from (77.2-88.9%). The classifiers identify each of the subjects with any color having an accuracy ranged from (70.9-88.9%), and the RFC shows the best accuracy which is 88.9% in the case of blue color stimuli.

Distraction descriptor for brainprint authentication modelling using probability-based Incremental Fuzzy-Rough Nearest Neighbour

“…For instance, when the stored brainprint of a client is stolen, disclosed, or lost, a different brainprint can be generated from a specific type of brain activities and responses. For example, the EEG signals can be recorded from different colour stimuli [ 4 ] or black and white stimuli [ 5 ]. Thus, a new brainprint could be used to substitute the stolen one.…”