Thus, by accurately predicting RT, the mental state can be more accurately assessed, yielding a better BCI accuracy. One of the pervasive challenges in BCI research is the reduced BCI accuracy over long sessions due to fluctuations in mental state. Most BCI applications favor EEG as the tool to infer the mental state and to understand communicative intent. Our study not only is useful to understand psychological phenomenology but also helps in the translation of brain signals into machine-comprehensible commands that can facilitate augmentative and alternative communication (AAC) using the brain–computer interface (BCI). This paper focuses on analyzing human response towards visual stimulus and provides methods of estimating RT from information embedded in electroencephalogram (EEG) signals. Accurate prediction of reaction time (RT) using neurophysiological biomarkers can help detect various mental states and develop better human–computer interfaces for patients and healthy subjects. Cognitive and affective state monitoring has become a topic of interest in understanding various sensory-motor functions. Thanks to advancements in sensor and hardware capabilities, and signal processing techniques, we now have better tools to understand the brain. Investigating further, we found that the left central as well as parietal and occipital lobes were crucial for predicting RT, with significant activities in the theta and alpha frequency bands. The regression-based approach predicted RTs with correlation coefficients (CC) of 0.78 and 0.80 for FCNN and CNN, respectively. With the FCNN model, the accuracies obtained for binary and 3-class classification were 93% and 76%, respectively, which further improved with the use of CNN (94% and 78%, respectively).
Though deep neural networks are widely known for classification applications, cascading FCNN/CNN with the Random Forest model, we designed a robust regression-based estimator to predict RT.
We have designed a Fully Connected Neural Network (FCNN) and a Convolutional Neural Network (CNN) to predict and classify RTs for each trial. In this paper, to the best of our knowledge, we introduce a first-ever DNN-based generalized approach to estimate reaction time (RT) using the periodogram representation of single-trial EEG in a visual stimulus-response experiment with 48 participants. Multiplexed deep neural networks (DNN) have engendered high-performance predictive models gaining popularity for decoding brain waves, extensively collected in the form of electroencephalogram (EEG) signals.
0 kommentar(er)
