Classification of fNIRS Data Under Uncertainty: A Bayesian Neural Network Approach

TL;DR

This paper employs a Bayesian Neural Network with Variational Inference to classify fNIRS brain signals, effectively accounting for data and model uncertainty, achieving high accuracy and providing insights into uncertainty quantification.

Contribution

It introduces a Bayesian Neural Network approach for fNIRS data classification that considers uncertainty, which is novel compared to traditional methods.

Findings

Achieved 86.44% accuracy on a binary classification task.

Demonstrated the convergence of the evidence lower bound (ELBO).

Produced a ROC curve with an AUC of 0.855.

Abstract

Functional Near-Infrared Spectroscopy (fNIRS) is a non-invasive form of Brain-Computer Interface (BCI). It is used for the imaging of brain hemodynamics and has gained popularity due to the certain pros it poses over other similar technologies. The overall functionalities encompass the capture, processing and classification of brain signals. Since hemodynamic responses are contaminated by physiological noises, several methods have been implemented in the past literature to classify the responses in focus from the unwanted ones. However, the methods, thus far does not take into consideration the uncertainty in the data or model parameters. In this paper, we use a Bayesian Neural Network (BNN) to carry out a binary classification on an open-access dataset, consisting of unilateral finger tapping (left- and right-hand finger tapping). A BNN uses Bayesian statistics to assign a probability distribution to the network weights instead of a point estimate. In this way, it takes data and model uncertainty into consideration while carrying out the classification. We used Variational Inference (VI) to train our model. Our model produced an overall classification accuracy of 86.44% over 30 volunteers. We illustrated how the evidence lower bound (ELBO) function of the model converges over iterations. We further illustrated the uncertainty that is inherent during the sampling of the posterior distribution of the weights. We also generated a ROC curve for our BNN classifier using test data from a single volunteer and our model has an AUC score of 0.855.