Jupitara Hazarika

Objective. Multimodal neuroimaging fusion has shown promise in enhancing brain–computer interface (BCI) performance by capturing complementary neural dynamics. However, most existing fusion frameworks inadequately model the temporal asynchrony and adaptive fusion between electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS), thereby limiting their ability to generalize across sessions and subjects. This work aims to develop an adaptive fusion framework that effectively aligns and integrates EEG and fNIRS representations to improve cross-session and cross-subject generalization in BCI applications. Approach . To address this, we propose STeCANet, a novel Spatiotemporal Cross-Attention Network that integrates EEG and fNIRS signals through hierarchical attention-based alignment. The model leverages fNIRS-guided spatial attention, EEG-fNIRS temporal alignment, adaptive fusion, and adversarial training to ensure robust cross-modal interaction and spatiotemporal consistency. Main results . Evaluations across three cognitive paradigms, namely motor imagery, mental arithmetic, and word generation, demonstrate that STeCANet significantly outperforms unimodal and recent multimodal baselines under both session-independent and subject-independent settings. Ablation studies confirm the contribution of each sub-module and loss function, including the domain adaptation component, in boosting classification accuracy and robustness. Significance . These results suggest that STeCANet offers a robust and interpretable solution for next-generation BCI applications.

Parkinson's disease (PD) is a neurological disorder which is progressive in nature. Although there is no cure to this disease, symptomatic treatments are available. These treatments can slow the progressive development of the symptoms. Medications can treat some of the symptoms of the PD up to a great extent that in turn may help the patients to live a normal life. Besides these medications, the patients can also be provided with various therapies based on the types of their symptoms. But for providing any treatment, detection of its symptoms at an early stage is very important. This can reduce its future complexities considerably. Early diagnosis along with proper medications may treat the symptoms of PD significantly. This motivates to propose a new and effective methodology for detection and analysis of PD. In this work, an approach has been proposed for identification of PD patients by using Electroencephalogram (EEG) signals. Here, the EEG signals of normal persons and PD patients are processed in three stages. First, the raw EEG signals are pre-processed for removal of noises and artefacts present. Out of various techniques, Wavelet transform is used for this purpose. In the MATLAB environment, de-noising can be executed by using the in-built functions. Performances of the de-noising techniques are examined with the performance parameters namely Root Mean Square Error (RMSE) as well as Signal to Noise Ratio (SNR). In the second stage, statistical features are extracted from the pre-processed EEG signals. In this work, five statistical features are considered for performing the classification. In the final stage, the extracted features are classified using Artificial Neural Network (ANN) and Support Vector Machine (SVM) techniques. ANN is an efficient classifier that predicts the human brain's working manners. SVM on the other hand has been proven as one of the most prevailing classification algorithms that gives highly accurate and robust results. This is a novel approach of analyzing the performances of the classification techniques by evaluating the best performing feature. In both the classifiers accuracy, precision, sensitivity and specificity are calculated from the confusion matrix evaluated from the values of the statistical features. In ANN, results using six different training algorithms at different hidden layers are calculated and compared. This proves the training algorithm Levenberg-Marquardt back-propagation with hidden layer 20 as the best combination for performing the classification. From the results it is seen that both ANN and SVM classifiers provide significant classification accuracies of 94.7% and 96.5% respectively. Amongst the five considered features, Mean performs the best in terms of classification accuracy.

The Electroencephalogram (EEG) signals recorded during motor imagery (MI) tasks are often used as input in the brain-computer interface (BCI) applications. The adaptation of these EEG signals to control signals depend on the selection of an accurate feature that can provide good classification. The study compared various features of alpha frequency band adapting wavelet-based time-frequency analysis approach. The mean, variance, wavelet energy, Shannon entropy, log energy entropy, kurtosis, and skewness feature were extracted from the EEG signals recorded from symmetrical electrode locations of the motor cortex and their performance accuracies were evaluated using SVM and KNN classification algorithms. In results, the Shannon entropy showed the highest classification accuracy of 86.4 % with SVM classifier. Findings of this study establish Shannon entropy as a potential feature to use in EEG-based BCI analysis along with SVM classification algorithm.