Person ıdentıfıcatıon usıng functıonal near- ınfrared spectroscopy sıgnals usıng a fully connected deep neural network

Abstract

Key word and phrases: functional near-infrared spectroscopy (fNIRS), person identification, PCA, SVD, fully connected deep neural network, random forest. Abstract. In this study, we investigate the suitability of functional near-infrared spectroscopy signals (fNIRS) for person identification using data visualization and machine learning algorithms. We first applied two linear dimension reduction algorithms: Principle Component Analysis (PCA) and Singular Value Decomposition (SVD) in order to reduce the dimensionality of the fNIRS data. We then inspected the clustering of samples in a 2d space using a nonlinear projection algorithm. We observed with the SVD projection that the data integrity associated with each person is high in the reduced space. In the light of these observations, we implemented a random forest algorithm as a baseline model and a fully connected deep neural network (FCDNN) as the primary model to identify person from their brain signals. We obtained %85.16 accuracy with our FCDNN model using SVD reduction. Our results are in parallel with the neuroscience researches, which state that brain signals of each person are unique and can be used to identify a person.

Enhanced Accuracy for Multiclass Mental Workload Detection Using Long Short-Term Memory for Brain–Computer Interface

“…In various brain imaging studies, fNIRS is used to investigate the hemodynamic activities and cognitive states such as MWL, vigilance, fatigue, and stress levels (Cain, 2007;Herff et al, 2013;Ho et al, 2019). Owing to the optical nature of fNIRS, the methodology is less prone to artifacts like a heartbeat or motor, head, and eye movements, which makes it the prevalent choice over other neuroimaging modalities like EEG, PET, and fMRI (Ozge Mercanoglu et al, 2017). The primary aim of this study was to explore the optimal ML or DL algorithms that best fit the four-phase MWL assessment and classification.…”

“…In another study, Huve et al (2018) repeated the same procedure for binary classification to control a robot. Hiwa et al (2016) and Ozge Mercanoglu et al (2017) attempted to predict the gender of the subjects through their unique fNIRS signals. Saadati et al (2019a,b) employed CNN using hybrid fNIRS-EEG settings for three-level MWL classification.…”

“…Hennrich et al (2015) and Naseer et al (2016) used DNN and other conventional classifiers to differentiate between two and three cognitive states using brain fNIRS signals. Some studies used similar procedures for binary classification to control robot and gender classification (Ozge Mercanoglu et al, 2017;Huve et al, 2018). Saadati et al (2019a,b) employed CNN with hybrid fNIRS-EEG for MWL classification and neurofeedback.…”