Background/Objectives: Advancements in neuroscience and machine learning have increasingly enabled brain pattern recognition based on bio-signal measurements, such as electroencephalography (EEG). These developments support next-generation technologies, including brain–computer interfaces (BCIs) and AI-assisted systems. However, volume conduction (VC) effects remain a major source of contamination in EEG recordings, affecting both univariate analyses and functional connectivity estimation. Methods: In this work, we propose a VC mitigation method that explicitly models and suppresses VC components in the observed functional networks. Specifically, the observed functional network is decomposed into a matrix capturing only VC-related components (i.e., components attributed to volume conduction) and a residual matrix, where the residual is regarded as a proxy for a VC-mitigated functional network that better reflects the underlying functional interactions. The VC component matrix is modeled using a decay function parameterized by the inter-electrode distance matrix, capturing the dominant spatial bias induced by VC. To estimate these parameters, we introduce supervised channel importance, quantified as the mutual information between experimental labels and channel signals, as a proxy for task-relevant neural activity. The parameters are optimized such that the unsupervised node importance derived from the VC-mitigated functional network, defined as the average node strength, aligns with the supervised channel importance. Results: Evaluation results using a deep-learning framework demonstrate that, compared with the observed functional network, the VC-mitigated functional network improves classification performance in brain pattern recognition tasks.

Advancements in neuroscience, especially in estimation technique of functional connectivity and brain networks, have been instrumental in emotion classification. However, conventional methods that utilize feature vectors or connectivity matrices do not entirely harness the spatial information of brain activity. We propose the Sliding Functional Connectivity Convolution (SFCC), which constructs brain networks by leveraging the spatial information of channel distribution of electroencephalogram (EEG). Similar to image convolution, SFCC begins by defining a matrix to be processed and an operation kernel, both derived from a channel projection. We then calculate the functional connectivity between the matrix elements and elements of the kernel while sliding the kernel across the matrix. The resultant brain networks provide additional spatial information for a classifier compared to conventional methods. Comparison between brain networks constructed using conventional feature vectors and connectivity matrices with those created by the proposed method showed a significant enhancement in classification performance.

Background/Objectives: This study proposes Diffusion Inverse Filtering (DIF), a spatially informed transformation designed to counteract spatial smoothing in functional-connectivity representations (i.e., functional networks) and thereby enhance their discriminative power for pattern recognition. Spatial smoothing in electroencephalography (EEG) signals and derived features, such as functional connectivity, is largely attributed to volume conduction. Functional connectivity has been increasingly used in brain–computer interface (BCI) studies; however, this spatial smoothing can introduce spurious connections and distort functional-connectivity patterns. Methods: DIF approximates spatial smoothing in functional connectivity, which is potentially associated with volume conduction, as a diffusion-like process and applies a regularized inverse operation to transform the observed functional networks into networks with enhanced discriminative representations. The effectiveness of DIF in enhancing the discriminative power of functional-connectivity representations in pattern recognition was evaluated using emotion recognition as the paradigm task, which is a critical component of BCI systems. Experimental results show that DIF generally improves emotion-recognition performance relative to the originally observed functional networks under electrode-sparsification conditions, with the most consistent improvements observed for Pearson correlation coefficient (PCC) estimation. Both signal-level processing, which operates on EEG signals before functional-connectivity estimation, and function-al-connectivity-level transformations, including graph signal processing (GSP)-based filtering applied after functional-connectivity estimation, were included as comparators. Under this framework, DIF is also considered a functional-connectivity-level transformation, but does not rely on a GSP framework. Results: The performance of DIF demonstrates the potential of functional-connectivity-level transformation as a complement or alternative to signal-level processing for enhancing connectivity-based pattern recognition. Overall, DIF improves classification performance and offers strong compatibility with modern functional-connectivity-based BCI pipelines.

Advancements in neuroscience and deep learning have significantly enhanced bio-signal-based emotion recognition, a critical component in Brain-Machine Interface (BMI) applications for healthcare, human-computer interaction, and human-AI assistant communication. Former studies have proposed Manual Mapping electrode matrices and employing Convolutional Neural Networks (CNNs) to recognize spatial EEG activities. However, this Manual Mapping of EEG electrodes onto matrix grids limits spatial precision and introduces inefficiencies. This study proposes automated channel mapping methods of Orthographic Projection and Stereographic Projection to address these challenges, using Differential Entropy and Power Spectral Density with Linear Dynamical Systems as features. A 3-branch multiscale CNN was trained on open-source dataset, employing a 5-fold cross-classification approach. Experimental results demonstrate that higher-resolution grids (16×16, 24×24) with automated projections significantly outperform Manual Mappings, achieving up to a 4.06% improvement in classification accuracy (p < 0.05). This result indicates that enhancing spatial precision of EEG data improves emotion classification, establishing automated spatial mapping as an advancement in EEG-based emotion recognition.Clinical Relevance—Advancement in emotion classification accuracy can facilitate more reliable diagnostic tools and personalized therapeutic interventions for mental health disorders, such as depression and anxiety.