Novel fNIRS study on homogeneous symmetric feature-based transfer learning for brain–computer interface

Abstract

The brain–computer interface (BCI) provides an alternate means of communication between the brain and external devices by recognizing the brain activities and translating them into external commands. The functional Near-Infrared Spectroscopy (fNIRS) is becoming popular as a non-invasive modality for brain activity detection. The recent trends show that deep learning has significantly enhanced the performance of the BCI systems. But the inherent bottleneck for deep learning (in the domain of BCI) is the requirement of the vast amount of training data, lengthy recalibrating time, and expensive computational resources for training deep networks. Building a high-quality, large-scale annotated dataset for deep learning-based BCI systems is exceptionally tedious, complex, and expensive. This study investigates the novel application of transfer learning for fNIRS-based BCI to solve three objective functions (concerns), i.e., the problem of insufficient training data, reduced training time, and increased accuracy. We applied symmetric homogeneous feature-based transfer learning on convolutional neural network (CNN) designed explicitly for fNIRS data collected from twenty-six (26) participants performing the n-back task. The results suggested that the proposed method achieves the maximum saturated accuracy sooner and outperformed the traditional CNN model on averaged accuracy by 25.58% in the exact duration of training time, reducing the training time, recalibrating time, and computational resources.

Four-Class Classification of Neuropsychiatric Disorders by Use of Functional Near-Infrared Spectroscopy Derived Biomarkers

“…The major advantages of these techniques rely on their ability to capture the complexity of neural information embedded in the HBO signal patterns through optimization of the network structures [ 81 ]. Indeed, there exists some successfully implemented DL classifiers with fNIRS and EEG signals [ 82 , 83 , 84 , 85 , 86 ]. However, we avoided testing the utility of DL algorithms in the presented work because of the limited cohort size of each group.…”

Deep learning in fNIRS: a review

“…Using the same dataset, Kwak et al 72 applied a branched CNN architecture that used the fNIRS data to generate spatial feature maps, which were then fed to the EEG maps to try to obtain higher spatial resolution than ordinary EEG and higher temporal resolution than fNIRS. The resulting classification accuracies were 78.97% for motor imagery and 91.96% for mental arithmetic tasks, which are only slight improvements over the tensor fusion methods of Sun et al Khalil et al 73 also used a fusion of fNIRS and EEG data to distinguish between rest and a mental workload tasks. With a CNN, an accuracy of 68.94% was achieved when trained on data from 5 of the 26 participants.…”

“…Khalil et al. 73 also used a fusion of fNIRS and EEG data to distinguish between rest and a mental workload tasks. With a CNN, an accuracy of 68.94% was achieved when trained on data from 5 of the 26 participants.…”

A Machine Learning Perspective on fNIRS Signal Quality Control Approaches

“…Regarding the application of deep learning approaches to fNIRS signals, only one study addressed the classification of the signal quality [10], while other examples addressed task and gesture recognition for Brain-Computer-Interaction applications [29], [30]. The study of Gabrieli and colleagues [10] aimed at using a CNN based to classify the quality of 510 short fNIRS portions.…”