Chen Xu, Yu ji, Zhenyu Lv, Yang Yi, Yizhe Yang, Luyao Ji, Chaoyi Chen, Xianyang Wang, Tian Lan, Zhihua Wang, Juan Wang, Xunde Dong, Fuze Tian, Qunxi Dong, Bin Hu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Multimodal fusion of Electroencephalography (EEG) and functional Near-Infrared Spectroscopy (fNIRS) has shown great promise in Brain-Computer Interface (BCI) tasks. However, due to differences in physical mechanisms, temporal dynamics, and semantic representations between the two modalities, the fusion process faces significant challenges such as heterogeneity and temporal misalignment. To address this, we propose a cross-modal decoupled deformable distillation (CMD$^{3}$) method, which aims to achieve flexible, efficient, and interpretable EEG-fNIRS fusion learning. CMD$^{3}$first decouples the feature representations of each modality into modality-independent and modality-specific spaces to separately model commonality and complementary information. A deformable feature extraction network is then designed to process shared and specific features individually, enabling cross-modal temporal alignment via predicted dynamic offsets, thereby mitigating response delays between modalities. Furthermore, to facilitate inter-modal knowledge transfer, we construct a dual-space graph distillation module to explicitly migrate semantic information across modalities, with learnable edge weights used to adaptively regulate the distillation strength. CMD$^{3}$is systematically evaluated on public datasets covering emotion recognition and motor imagery tasks. Experimental results demonstrate that CMD$^{3}$consistently outperforms existing fusion approaches in classification performance. Offset visualization further reveals physiologically meaningful temporal attention patterns learned by the model, validating the effectiveness and explainability of the proposed method.

Electroencephalogram (EEG) has been widely used for emotion recognition due to its portability and high temporal resolution. It makes success in subject-dependent scenario but faces significant challenges in cross-subject emotion recognition because of non-stationarity of EEG and individual differences. Most previous studies treat all individuals as a single source domain for transferring emotional knowledge, which may introduce irrelevant information and lead to negative transfer. Besides, there is a potential risk that some important information of common emotional features might be ignored. To deal with the issues, we propose a framework called hybrid source selection fusion domain-invariant attention (HSSFDA) for cross-subject emotion recognition. First, source domains are selected by leveraging local and global similarity for knowledge transfer. Then, a specialized attention mechanism is employed to focus on important emotional information extracted from the domain-invariant features. Finally, domain-invariant and domain-specific features are fused to enhance emotion recognition performance. To evaluate the proposed method, experiments are conducted on several public datasets including SEED, SEED_IV, DREAMER and DEAP. The results demonstrate that HSSFDA achieves accuracies of 85.07 %, 72.11 %, 62.36 %, 77.17 %, 58.51 %, and 63.55 % on SEED, SEED_IV, valence and arousal of DREAMER, and valence and arousal of DEAP datasets, respectively, demonstrating competitive performance compared to popular and state-of-the-art methods. Furthermore, we apply the HSSFDA to a self-recorded dataset collected by self-developed three-channel device and validate its effectiveness in practical applications. In conclusion, HSSFDA is a feasible method for cross-subject emotion recognition and has the potential to broaden the application of EEG in the field of affective computing.

Researchers have long sought objective and quantifiable methods for recognizing depression. Electroencephalography (EEG) signals, which reflect brain activities objectively, have emerged as a promising tool for this purpose. However, the practical application of EEG signals faces significant challenges arising from distribution variability across different datasets and subjects. In addition, conventional methods often struggle to effectively capture information related to dynamic transformations in distributions. To address these issues, we propose a Wasserstein distribution-inspired dynamic adversarial network (WDANet) for EEG-based depression recognition. Specifically, WDANet includes a global discriminator that focuses on the marginal distribution of EEG features, a local discriminator that concentrates on the conditional distribution of EEG features, and a Wasserstein distribution discriminator that utilizes Wasserstein distributions derived from various processed EEG features. The experimental results show that WDANet achieved classification accuracies of 83.33%, 75.52%, 73.93%, 76.04%, and 70.94% in cross-subject, cross-dataset experiments conducted on three datasets, demonstrating its effectiveness and superiority compared to state-of-the-art methods. These results support our claim that WDANet enhances the accuracy and interpretability of depression recognition, providing insights and new research directions for the integration of neuroscience and artificial intelligence technologies.

Depression is a common emotional disorder in modern society that causes growing burdens globally. This mental disorder has been frequently confirmed to be closely related to abnormalities in the prefrontal cortex (PFC). However, it remains to be fully investigated whether the neural activity in the PFC has regular quasi-steady spatiotemporal structures, and whether dynamic patterns of these structures are associated with prefrontal dysfunction in depression. To further uncover such neural correlates, we extended the traditional electroencephalography (EEG) microstates to a novel localized level and developed a variational Bayesian probabilistic generative model to decode such localized microstate patterns from few-channel prefrontal EEG signals. We validated the method with a publicly available multichannel EEG dataset obtained from 165 healthy individuals and a three-channel prefrontal EEG dataset (43 depressed and 43 healthy). The approach was finally used to examine dynamic evolution of prefrontal neural activities in depression. Our results demonstrate that the localized prefrontal microstates exhibit high cross-dataset reproducibility and were characterized by finer spatiotemporal patterns independent from traditional whole-brain microstates. The results further revealed significant emotional task-specific abnormalities in localized prefrontal microstates between the depressed and the healthy individuals, including more frequent occurrences and shorter durations in high-power bilaterally asymmetric microstates, as well as less organized low-power symmetric microstates. Our extended concept of the localized microstates and associated analytical methods provides a novel theoretical framework for elucidating prefrontal neural dynamics and also lays a theoretical foundation for uncovering prefrontal functional abnormalities in depression and developing auxiliary diagnostic tools with prefrontal few-channel EEG data.

Multimodal peripheral physiological signal fusion for emotion recognition seeks to perceive or recognize human emotions using peripheral modalities such as electromyography, electrodermal activity, and respiratory wave. Previous approaches to multimodal fusion primarily focus on emotion-sensitive signals such as electroencephalogram (EEG), often overlooking the potential value of peripheral physiological signals in emotion recognition. Moreover, the inherent heterogeneity among different modalities continues to pose challenges to fusion quality. In this article, we propose a multimodal decoupled multimodal fusion (DMF) to address these issues, enabling flexible feature decoupling, cross-modal feature interaction, and relational knowledge learning. Specifically, each modality’s representation is first decoupled into two components: common features and modality-specific features; second, the DMF employs progressive cross attention to facilitate the exchange of modality-specific features across different modalities; and finally, it uses relational knowledge to learn multimodal spliced features, embedding both inter-modal and intra-modal feature relationships. DMF offers a dynamic multimodal emotion recognition framework that leverages the emotional information contained in diverse modalities. Experimental results demonstrate that the DMF method consistently outperforms previous approaches and provides a viable solution for multimodal peripheral physiological signal fusion.

Event-driven neuromorphic computing involves sparse and asynchronous signal activity, which leads to irregular computation patterns and fine-grained concurrency. As a result, processing architectures need to support both high parallelism and energy efficiency. Among existing architectural solutions, superscalar designs exhibit significant potential for addressing high parallelism demands. However, conventional superscalar processors, which rely on synchronous circuits, maintain high-frequency clocking at all times, leading to substantial power inefficiency in sparse computation scenarios. To address this issue, we propose an asynchronous superscalar architecture that replaces global clocking with fully local handshake-based control, implemented using a bundled-data asynchronous protocol. The design supports decoding of up to 64 scalar instructions per cycle and implements the RISC-V RV32IMC instruction set. A prototype was fabricated using a 110 nm complementary metal oxide semiconductor (CMOS) process and was evaluated through post-layout simulation. Operating at 1.2 V, the processor delivers a peak INT8 throughput of 669.4 GOPS, with a static power consumption of 421 mW.

Depression detection using Electroencephalogram (EEG) signals obtained from wearable medical-assisted diagnostic systems has become a well-established approach in the field of affective disorders. However, despite recent advancements, on-board Artificial Intelligence (AI) models still demand substantial computational resources, presenting significant challenges for deployment on resource-constrained wearable medical devices. Embedded Multi-core Processors (MPs) offer a promising solution for accelerating these models. However, the limited computational capabilities of embedded MPs, combined with the structural diversity of AI models, complicate resource allocation and increase associated costs. To address these challenges, we propose a Memory-Aware Multi-Objective Iterative Local Search (MAMILS) algorithm to optimize task scheduling, thereby improving the efficiency of AI model deployment on wearable EEG devices. Experimental results across seven AI models demonstrate that, the MAMILS approach yields substantial improvements in key performance indicators: Total Energy Consumption ($\bm {TEC}$) with an average reduction of 47.57%,$\bm {Makespan}$with an average reduction of 48.75%, and$\bm {Throughput}$with an average increase of 198.37%, all while maintaining satisfactory classification performance for both Machine Learning (ML) and Deep Learning (DL) models. Especially, on-board deployment of EEGNeX achieves an accuracy of 93.4%, sensitivity of 91.6%, and specificity of 95.8%. Further analysis indicates that, when integrated with wearable EEG sensors and executable on-board AI models, the proposed MAMILS optimization strategy shows significant promise in facilitating the widespread adoption of low-power, real-time diagnostic systems for depression detection.

Prolonged mental fatigue poses significant risks to both individual health and societal productivity. Traditional methods of detecting mental fatigue often neglect the importance of timely, accessible monitoring. Although blue-enriched light (BEL) is effective at mitigating mental fatigue, prolonged exposure to it can result in visual fatigue and impairment. To address these limitations, we propose a portable fatigue monitoring and BEL feedback system. The system uses three-lead prefrontal Electroencephalogram (EEG) signals to detect mental fatigue in real time and dynamically adjusts BEL exposure based on the user's current fatigue state. This prevents visual issues associated with prolonged BEL exposure. In this work, we develop an onboard executable Convolutional Neural Network (CNN) model using interpretable two-dimensional (2D) convolution implemented with TensorFlow Lite. Our model strikes a favorable balance between classification accuracy (85 %) and computational efficiency, requiring only 50.98 K Floating-Point Operations (FLOPs) and a parameter size of 7.28 KB. When deployed on an EEG sensor, the model operates with just 79.98 KB of RAM and 378.98 KB of ROM. It performs a single inference in 130 ms and consumes 232.31 mW of power per inference. Experimental results demonstrate that the lightweight, onboard executable model integrated with the custom-designed hardware system shows potential for effectively modulating mental fatigue.

Music therapy has been shown to be effective in treating depression, as supported by numerous clinical studies and randomized controlled trials utilizing subjective reports and psychometric scales. However, a more efficient and objective approach is needed to complement these assessments and to investigate the neural effects of music therapy. This study designed a validity assessment framework based on the dot-probe paradigm, which uses participants' attentional biases toward emotional faces as an objective marker of music therapy effectiveness. To validate this framework, we collected 64 -channel electroencephalogram (EEG) signals from patients undergoing music therapy during a dot-probe task, and explored the dynamic cognitive processes in depressed patients before and after treatment using event-related potentials (ERPs). Our results indicate modulated attentional biases toward negative emotional faces in depressed patients following music therapy. Specifically, we observed shorter response times and higher ERP amplitudes for negative faces compared to positive faces before treatment. These abnormalities were ameliorated after treatment and showed significant correlations with pre- and post-treatment scale measurements. These findings lay a foundation for future artificial intelligence-based systems that could automate the assessment of music therapy effectiveness using neurophysiological markers, potentially enabling personalized treatment approaches and real-time therapeutic adjustments.

Heroin addiction represents a chronic neuropsychiatric disorder characterized by profound alterations in the brain's reward circuitry, particularly within the medial prefrontal cortex (mPFC), ventral tegmental area (VTA), and nucleus accumbens (NAc). Despite mounting evidence that addiction involves dysregulated neural oscillations, frequency domain analysis of reward processing in deeper brain structures remains critically understudied. We collected$\mathbf{6 4}$-channel EEG data from heroin addicts and healthy controls during a monetary incentive delay task with three conditions: positive (potential monetary gain), neutral (no gain or loss), and negative (potential monetary loss). Using advanced source localization techniques with fine realistic head models, we reconstructed brain source signals from key reward circuitry regions (mPFC, VTA, and NAc), analyzing neural responses across three temporal stages: reward anticipation, reward expectation, and reward outcome through event-related desynchronization/synchronization (ERD/ERS) analysis and time-frequency analysis. Heroin addicts exhibited significantly altered neural oscillatory patterns compared to healthy controls across delta, alpha, beta, and gamma frequency bands within the reward circuitry, with frequency-specific abnormalities observed in both cortical and subcortical reward-related regions during different stages of reward processing. This study provides the first comprehensive characterization of frequency-specific neural dysfunction spanning the entire reward circuitry in heroin addiction, offering novel insights into the oscillatory mechanisms underlying reward processing abnormalities and informing the development of frequency-targeted therapeutic interventions.

Recently, the electroencephalogram (EEG) has been widely adopted as a quantitative indicator for monitoring emotional modulation during virtual reality (VR) experiences. Although VR emotion-induction materials are continuously being developed, few methods have been proposed for constructing VR scenarios through psychophysiological calibration. To achieve precise emotional regulation, we propose an integrated psychophysiological oriented EEG-based VR scenario modeling approach for emotion induction. This methodology constructs 3D VR scenarios by extending 2D elements through core processes: (1) deconstruction of emotion-annotated 2D images to extract visual-audio emotional patterns, (2) development of immersive environments with dynamic camera trajectories, and (3) integration of audio stimuli and real-world physical configurations to form multi-sensory emotional stimuli. During emotioninduction experiments, EEG signals reflecting emotional states were captured and analyzed for each scenario. We implemented multidimensional calibration of emotion-induction materials by: (a) calculating emotion indicators from EEG signals, (b) combining these with traditional assessments using the Self-Assessment Manikin (SAM) scale, and (c) calibrating scenarios for distinct emotion categories. This yields psychophysiologically calibrated VR scenarios for emotion induction, delivering standardised affective stimuli with synchronised EEG datasets to advance affective computing research.

Colorectal Cancer (CRC) is one of the leading causes of cancer-related deaths worldwide, and early screening plays a crucial role in improving patient outcomes. In this study, we present a novel AI-assisted CRC diagnostic system using Bowel Sound (BS) signals. We first develop two portable BS acquisition devices with distinct form factors for high-fidelity signal capture in both clinical and home-care scenarios. A total of 221 recordings were collected under expert-guided protocol, with 144 CRC recordings and 59 Non-CRC healthy controls using the developed device. To enable low-resource deployment, we design a lightweight deep learning model optimized for real-time, on-board inference. The model incorporates multiple training strategies, including transfer learning on a large-scale public BS dataset, self-supervised temporal feature learning, and a hybrid semi-and weakly-supervised approach that leverages both unlabeled and real-noise data. Furthermore, a Sound Event Detection (SED) attention mechanism and iterative consistency learning are introduced to enhance the model’s sensitivity to BS activity. The proposed model comprises only 264.7 K parameters and 253.2 M Floating-Point Operations (FLOPs), requiring 1.57 MB of RAM and 1.03 MB of FLASH when deployed on microcontroller. It performs inference in approximately 3.4 s with low power consumption, making it well-suited for low-resource environments. Despite its compact design, the model achieves 93.06% classification accuracy, 96.46% sensitivity, and 86.99% specificity for binary-classes in CRC diagnosis. These results demonstrate the system’s potential for accessible and cost-effective CRC screening in community, home, and rural healthcare scenarios.

In noninvasive brain-computer interfaces (BCIs), EEG analysis plays a critical role, with neural networks serving as a cornerstone for signal decoding. Existing neural network approaches for EEG signal recognition require extensive manual design and hyperparameter tuning, leading to inefficiencies and making them impractical for embedded devices due to their large model size. To address these limitations, we propose mixed-level differentiable and hardware-aware neural architecture search (MDH-NAS), a framework that automatically generates lightweight neural networks tailored for EEG classification. Unlike traditional DARTS methods, MDH-NAS employs a hybrid optimization strategy that balances global and local search spaces, thereby accelerating and refining architecture discovery. It introduces explicit size constraints during the search process to ensure deployability on embedded devices. MDH-NAS demonstrates autonomous generation of architectures for tasks such as motor imagery (MI) and depression recognition, achieving 87.80% accuracy on the BCI-IV dataset and 90.09% on the MODMA dataset. When deployed on the EAIDK-610 board across heterogeneous tasks, it attains 85.37% accuracy on the EEG Motor Movement/Imagery dataset. This method reduces architecture discovery time by 89% and enhances prediction accuracy by 8.70% compared to baseline methods, highlighting its potential for scalable EEG analysis and real-world embedded deployment.

Feature selection is a crucial step in EEG emotion recognition. However, it was often used as a single objective problem to either reduce the number of features or maximize classification accuracy, while neglecting their balance. To address the issue, we proposed Improved Multi-objective Grey Wolf Optimization Feature Selection (IMGWOFS). First, we designed a population initialization operator via discriminability and independence of features to accelerate search speed. Second, we employed a two-stage update strategy to improve the global search capabilities of the EEG feature subsets. Finally, we incorporated an adaptive mutation operator to escape the local optima. We conducted experiments on SEED and DEAP datasets, and the accuracy were 86.87$\pm$1.62 % and 60.65$\pm$1.51 % in the beta band using a smaller number of EEG features. In addition, the frontal lobe was related to emotion processing. In conclusion, IMGWOFS is an effective and feasible feature selection method for EEG-based emotion recognition.

Mental disorders have increased rapidly and have emerged as a serious social health issue in the recent decade. Undoubtedly, the timely treatment of mental disorders is crucial. Emotion regulation has been proven to be an effective method for treating mental disorders. Music therapy as one of the methods that can achieve emotional regulation has gained increasing attention in the field of mental disorder treatment. However, traditional music therapy methods still face some unresolved issues, such as the lack of real-time capability and the inability to form closed-loop systems. With the advancement of artificial intelligence (AI), especially AI-generated content (AIGC), AI-based music therapy holds promise in addressing these issues. In this paper, an AIGC-based closed-loop music intervention system demonstration is proposed to regulate emotions for mental disorder treatment. This system demonstration consists of an emotion recognition model and a music generation model. The emotion recognition model can assess mental states, while the music generation model generates the corresponding emotional music for regulation. The system continuously performs recognition and regulation, thus forming a closed-loop process. In the experiment, we first conduct experiments on both the emotion recognition model and the music generation model to validate the accuracy of the recognition model and the music quality generated by the music generation models. In conclusion, we conducted comprehensive tests on the entire system to verify its feasibility and effectiveness.

The advancement of wearable electronic technology has facilitated the integration of smart wearable devices into artificial intelligence (AI)-driven medical assisted diagnosis. Embedded multi-core processors (MPs) have gradually emerged as pivotal hardware components for smart wearable medical diagnostic devices due to their high performance and flexibility. However, embedded MPs face the challenge of balancing performance, power consumption, and load-balancing. In response, we introduce a Pareto-based iterated local search (PILS) algorithm for task scheduling, which systematically optimizes multiple objectives, alongside a task list model to reduce the dimension of the decision space and enhance scheduling performance. In addition, we present a two-stage discretization scheme to ensure that the proposed algorithm offers meaningful guidance throughout the scheduling process. Simulation and on-board testing results show that the proposed algorithm effectively optimizes energy consumption, task execution time, and load balancing in embedded MPs task scheduling, indicating the potential of the proposed algorithm in enhancing the performance of smart wearable medical diagnostic devices powered by embedded MPs.

Emotions regulation play an important role in human behavior, but exhibit considerable heterogeneity among individuals, which attenuates the generalization ability of emotion models. In this work, we aim to achieve robust emotion prediction through efficient disentanglement of affective semantic representations. In detail, the data generation mechanism behind observations from different perspectives is causally set, where latent variables that relate to emotion are explicitly separate into three parts: the intrinsic-related part, the extrinsic-related part, and the spurious-related part. Affective semantic features consist of the first two parts, with the understanding that spurious latent variables generate the inherent biases in the data. Furthermore, a variational autoencoder with a reformulated objective function is proposed to learn such disentangled latent variables, and only adopts semantic representations to perform the final classification task, avoiding the interference of spurious variables. In addition, for electroencephalography (EEG) data used in this article, a space-frequency mapping method is introduced to improve information utilization. Comprehensive experiments on popular emotion datasets show that the proposed method can achieve competitive intersubject generalization performance. Our results highlight the potential of efficient latent representation disentanglement in addressing the complexity challenges of emotion recognition.

Emotion inhibitory control is a key executive function of the human brain, which regulates behavior by suppressing inappropriate responses. It plays an integral part in alleviating negative emotions, improving mood, and preventing depression. Transcutaneous auricular vagus nerve stimulation (taVNS) has been proved to enhance behavioral control, potentially suppressing negative emotions or facilitating their reduction in healthy individuals. However, the neurocomputational mechanisms underlying taVNS-induced neuroenhancement remain unclear. In this work, a portable electroencephalography (EEG) acquisition and stimulation device is designed to collect eight-channel EEG signals and deliver taVNS to both sides of ears. Then, we design a protocol that successfully induced negative emotions in healthy subjects. Next, we conduct a sham-controlled experiment, involving 28 healthy subjects, to explore the changes in EEG of negative emotions under taVNS. Finally, we primarily analyze the power spectrum density (PSD) of EEG signals and the functional connectivity network of the brain, based on the phase locking value (PLV), to assess the effect of taVNS on neural activity induced by negative emotions. The results of the experiment reveal that taVNS is a promising method for enhancing emotional inhibitory control by reducing PSD in the alpha band and enhancing PLV within prefrontal inhibitory control networks. In addition, differences in graph theory parameters between the Sham and taVNS conditions indicate that taVNS helps regulate negative emotions. In conclusion, this study demonstrates that taVNS enhances inhibitory control and reveals its neurocomputational mechanisms of EEG in healthy individuals during the development of negative emotions. And results indicate that taVNS could serve as a promising neuromodulation therapy for psychiatric disorders and individuals with depression or emotional distress.

Virtual reality (VR) mindfulness meditation has emerged as a prominent emotion regulation strategy in recent years. Current research often seeks to enhance meditation effectiveness through biofeedback and overlooks the trajectory of emotional changes and the changing needs during regulation. In this study, we propose an electroencephalography (EEG)–based positive feedback mechanism for VR mindfulness meditation aimed at optimizing the effects of emotion regulation. This mechanism consists of three modules: 1) EEG-based emotional state computation; 2) process-based relaxation assessment; and 3) adaptive positive decision feedback. Collectively, these components form a computation-assessment-feedback closed-loop system that objectively quantifies emotions while enabling real-time decision adjustments based on emotional trends, thereby enhancing user engagement and emotion regulation efficacy through personalized feedback. The contribution of the proposed feedback mechanism was evaluated through a randomized controlled trial (N= 36). The results indicated that both physiological measures and self-reported relaxation significantly increased when compared to interventions without feedback. These findings validate that the EEG-based positive feedback mechanism effectively enhances emotion regulation while providing additional insights into improving both the engagement and effectiveness within digital mental health interventions.

The development of affective computing and medical electronic technologies has led to the emergence of Artificial Intelligence (AI)-based methods for the early detection of depression. However, previous studies have often overlooked the necessity for the AI-assisted diagnosis system to be wearable and accessible in practical scenarios for depression recognition. In this work, we present an on-board executable multi-feature transfer-enhanced fusion model for our custom-designed wearable three-lead Electroencephalogram (EEG) sensor, based on EEG data collected from 73 depressed patients and 108 healthy controls. Experimental results show that the proposed model exhibits low-computational complexity (65.0 K parameters), promising Floating-Point Operations (FLOPs) performance (25.6 M), real-time processing (1.5 s/execution), and low power consumption (320.8 mW). Furthermore, it requires only 202.0 KB of Random Access Memory (RAM) and 279.6 KB of Read-Only Memory (ROM) when deployed on the EEG sensor. Despite its low computational and spatial complexity, the model achieves a notable classification accuracy of 95.2%, specificity of 94.0%, and sensitivity of 96.9% under independent test conditions. These results underscore the potential of deploying the model on the wearable three-lead EEG sensor for assisting in the diagnosis of depression.

Heart sounds play a crucial role in assessing Coronary Artery Disease (CAD). The advancement of Artificial Intelligence (AI) technologies has given rise to Computer Audition (CA)-based methods for CAD detection. However, previous research has focused primarily on analyzing and modeling heart sound data, overlooking practical application scenarios. In this work, we design a pervasive heart sound collection device used for high-quality heart sound data acquisition. Moreover, we introduce an on-board executable lightweight network tailored for the designed portable device, referred to as TYKDModel. Further, heart sound data from 41 CAD patients and 22 non-CAD healthy controls are collected using the developed device. Experimental results show that the TYKDModel exhibits low-computational complexity, with 52.16 K parameters and 5.03 M Floating-Point Operations (FLOPs). When deployed on the board, it requires only 1.10 MB of Random Access Memory (RAM) and 236.27 KB of Read-Only Memory (ROM), and takes around 1.72 seconds to perform a classification. Despite the low computational and spatial complexity, the TYKDModel achieves a notable classification accuracy of 85.2%, specificity of 88.6%, and sensitivity of 82.8% on the board. These results indicate the promising potential of AI-assisted all-in-one integrated system for the diagnosis of heart sound-assisted CAD.

Depression detection via wearable Electroencephalogram (EEG) sensor-assisted diagnosis system demands computationally efficient models compatible with resource-constrained edge devices. Spiking Neural Networks (SNNs) offer inherent advantages for processing the spatio-temporal patterns of EEG through event-driven manner. In this study, we innovatively present LSNNet, a lightweight SNN model specifically designed for wearable EEG sensors. The model exhibits low computational complexity with 7.18 K parameters and 67.68 M Floating-Point Operations (FLOPs). It requires only 246.88 KB of Random Access Memory (RAM) and 57.33 KB of Read-Only Memory (ROM) for on-board execution, and has been validated on both the single-core STM32U535CET6 and the multi-core GAP8 microcontrollers. Despite its minimal computational and memory requirements, LSNNet achieves impressive performance metrics, with a classification accuracy of 89.2%, specificity of 92.4%, and sensitivity of 86.4% in independent tests conducted on EEG data collected from 73 depressed patients and 108 healthy controls using our three-lead EEG sensor. Especially, when running on the GAP8 microcontrollers, the LSNNet model has a low power consumption of 21.43 mW and a satisfactory inference time of 0.63 s while maintaining a classification accuracy of 87.5% (only with a reduction of 1.98%). These results underscore the potential of integrating wearable EEG sensors with the LSNNet model for depression detection in the Internet of Things (IoT) era.

Sleep is a daily activity essential for well-being, yet many modern individuals experience sleep problems, and prolonged poor sleep negatively impacts both physiological and psychological health. Aromatherapy, an emerging complementary alternative medicine, has shown promise as a sleep aid, but it lacks objective and reliable monitoring and control methods. To address this gap, we propose a wearable, portable sleep monitoring and aromatherapy system that utilizes single-channel electroencephalogram (EEG) signals from the prefrontal lobe for sleep detection and provides aromatherapy feedback based on the detected sleep state. Our system incorporates a lightweight model based on convolutional neural networks (CNN) and long short-term memory (LSTM) networks, enabling on-board execution and classification. Trained on the publicly available Sleep-EDF dataset, the model achieves a classification accuracy of 85.1% and a macro-F1 score of 79.5%. The combination of our developed EEG sensor and the proposed model presents a promising solution for effective sleep monitoring and intervention, aiming to enhance sleep quality.

With great pride and enthusiasm, we present the inaugural edition of IEEE Transactions on Computational Social Systems (TCSS) for 2024. Reflecting on the year gone by, 2023 stands as a hallmark of academic excellence and prolific output, wherein our journal has successfully disseminated a substantial volume of scholarly work—301 articles encompassing approximately 3600 pages, distributed across six distinct issues.

Functional connectivity (FC) network derived from resting-state functional magnetic resonance imaging (rs-fMRI) has been extensively employed in the automated identification of brain disorders. Conventional FC network modeling methods typically assign equal importance to different sampling points of rs-fMRI and brain regions of interest (ROIs). Nevertheless, considering temporal and regional heterogeneity, this assumption may not always be applicable. Moreover, sparse regression algorithms with regularization terms are commonly adopted to eliminate spurious FC. However, these conventional regularization terms impose a uniform sparse penalty on all FC without considering the prior knowledge that the brain tends to transmit information in an energetically efficient manner. To address the above problems, we propose an adaptive weight and Wasserstein distance constrained low-rank sparse representation (AW-WD-LSR) method to construct FC networks. Specifically, we employ adaptive weights to the reconstruction errors of rs-fMRI time series across various ROIs and sampling points, thereby amplifying the significance of crucial features in the joint representation, leading to a more robust FC network. Meanwhile, we incorporate a local constraint by introducing the optimal transport distance (i.e., Wasserstein distance) between ROIs to adjust the sparse penalty on the corresponding FC. The larger Wasserstein distance, the higher information transmission cost between two ROIs, resulting in a greater penalty imposed on the corresponding FC. The efficacy of the innovation modules is demonstrated in classification tasks involving major depressive disorder (MDD) with mixed features (MMF), MDD without mixed features (MDDnoMF), and healthy controls (HCs).

Currently, the integration of artificial intelligence (AI) techniques with multimodal physiological signals represents a pivotal approach to detect affective disorders (ADs). With the increasing complexity and diversity of physiological signal modalities, researchers have introduced various AI methods using multimodal physiological signals to improve model classification performance and explainability to increase trust and facilitate clinical adoption. Among these methods, spiking neural networks (SNNs) stand out as a promising avenue due to their alignment with the operating principles of the human brain, robust biological explainability, and adeptness in processing spatial–temporal information in an efficient event-driven manner with low power consumption. Furthermore, the emergence of neuromorphic computing (NC) chips based on SNNs has greatly bolstered the field of NC, enabling effective support for objective, pervasive, and wearable AI-assisted medical diagnostic devices for ADs and other diseases. This article presents a review of recent achievements in multimodal AD detection and points out the associated challenges in utilizing multimodal physiological signals and NC based on SNNs for AD detection. Building upon this foundation, we give perspectives on future work. The intended readership for this review consists of researchers in the fields of cognitive computing, computational psychophysiology, affective computing, NC, and brain-inspired computing. We hope that this survey not only garners increased attention from the scientific community but also serves as a valuable guide for future studies in this field.

Mental fatigue is a prevalent issue in contemporary society and can negatively affect physical performance and concentration, increasing the likelihood of adverse consequences due to inattention during productive activities. Therefore, it becomes increasingly important to address and eliminate fatigue within a specific period of time. Aromatherapy, as a form of Complementary Alternative Medicine (CAM), is a non-invasive, cost-effective, and efficient method to combat fatigue. Previous studies have assessed the effects of specific aromatherapy oils using scales, but there is a lack of objective and reliable physiological indicators to prove the effectiveness of aromatherapy. Hence, this paper seeks to establish a model illustrating the effects of aromatic essential oil gases on the human body. A multimodal physiological fatigue signal acquisition system that integrates aromatherapy feedback was designed. In addition, an experimental paradigm was developed to explore the potential of aromatherapy in mitigating mental fatigue. Electroencephalogram (EEG) and Electrocardiogram (ECG) signals were collected, allowing for the analysis of time-frequency domain features in EEG and ECG signals, as well as Heart Rate Variability (HRV) features in ECG signals. Our findings indicate that specific aromatic gases demonstrate effectiveness in reducing mental fatigue. Furthermore, we employed the Support Vector Machine (SVM) algorithm to classify the state of human mental fatigue. Based on the classification results, the release of aromatic gas was controlled to provide targeted aromatic feedback. This innovative approach offers a promising avenue for objectively assessing and addressing mental fatigue through aromatherapy interventions.

In the Doppler biological radar-based applications of noncontact measurement of vital signs, effectively extracting heartbeat information from weak thoracic mechanical motion is an important problem to be solved. This study is aimed at extracting heartbeat signal via the technology combined with Short Time Fourier Transform (STFT), Singular Value Decomposition (SVD) and Adaptive Noise Canceller (ANC) from radar recording. The simulated data and the data collected by Doppler radar biosensor realized in laboratory are employed to validate the proposed method. The results show that the proposed method has the ability of detection for the heart rate and heart rate variability indexes in rest state, it has certain advantages in time-consuming and detection accuracy. Therefore, the current method provides another way to process vital sign signals recorded by Doppler radar.

Mental fatigue of the brain will cause the weakening of the psychological and physiological functions of the human body, which increases the risk of mistakes and accidents. This paper designs a mental fatigue detection and mitigation system based on human physiological signals and classical music. Through the fatigue assessment and mitigation experiment on 20 healthy subjects, it is verified that the system is able to judge the mental fatigue state of the subjects and play classical music to mitigate fatigue via acquiring and analyzing (alpha+theta)/beta of electroencephalogram (EEG) and heart rate variability (HRV) of ballistocardiogram (BCG). The designed system realizes signal acquisition and fatigue analysis only through using portable device, this makes it have certain application value in office building, hospital, aircraft driving and other scenes.

Sleep inertia is a transitional state from sleep to wakefulness, accompanied by groggy feelings and cognitive impairment. Previous research on sleep inertia mainly used expensive and cumbersome equipment, and the analysis of physiological signals relied on computers. This work introduces a sleep inertia detection system that consists of a wearable low-power electroencephalogram (EEG) acquisition module based on STM32WB55 and ADS1299, and a data processing module based on the Xilinx®Zynq®-7000 XC7Z020. This work recorded the EEG signals of ten subjects in the alert and sleep inertia states to extract the delta power, alpha power, beta power, EEG vigilance, and sample entropy. A linear support vector machine (SVM) was then used to classify the two states based on all subjects’ EEG signals, with an accuracy of 72.5%, and the average accuracy based on a single participant was 8S.9%. Finally, the feature extraction algorithm and SVM parameters were entered into the Zynq®system-on-chip (SoC) development board to realize onboard processing of the algorithm. The system is capable of evaluating the severity of human sleep inertia, which has reference significance for the practical application of sleep inertia detection.

Judging mental fatigue can be guided by acquiring and analyzing various physiological data. However, the existing equipment focuses on single physiological signals, such as electroencephalogram (EEG) and electrocardiogram (ECG), which ignores the preponderance of multimodal physiological signals in fatigue states. Alternatively, some equipment is difficult to operate and thus unsuitable for portable applications. To this end, this paper details the design of a miniaturized multi-physiological signal acquisition system. The system not only acquires EEG and ECG based on combining wavelet transform with Kalman filter, but also uses precise temperature sensors to synchronously acquire proximal skin temperature signals which are not easily interfered with by environmental noise or other physiological signals of human body. Through a fatigue assessment experiment on ten healthy subjects, it was verified that our equipment reliably detects mental fatigue states by monitoring and analyzing EEG, ECG, and proximal skin temperature data. In actual applications, this system appears to the traits of easy operation and good stability. It provides better hardware support for the pervasive applications and concrete implementation of mental health in multi-scene, and can popularize use.