More than two-thirds of dementia cases are attributed to Alzheimer’s disease (AD), while the remaining cases include frontotemporal dementia (FTD), vascular dementia, and other related disorders. Electroencephalography (EEG) is among the most cost-effective methods for supporting the diagnosis of these conditions and can serve as a valuable source of information for AI-assisted diagnostic systems. This paper focuses on the classification of EEG data from patients with FTD, Alzheimer’s disease, and healthy controls. Our study focused on two key issues in this setting. First, the reliable differentiation between FTD and AD. Secondly, EEG data are noisy, difficult to Pre-process, and often limited in their ability to capture long-range relationships. Modern Hopfield networks offer a promising direction because they are effective in pattern storage and retrieval and are closely related to attention mechanisms. In this work, four neural network architectures integrated with modern Hopfield networks are investigated on a publicly available dataset. A standardized workflow was adopted so that all models were trained and evaluated under identical conditions. The models were assessed using 5-fold stratified cross-validation together with hold-out evaluation. The best-performing model achieved 96% accuracy in the present experimental setting. Overall, the results show that the more expressive Hopfield-based architectures improve performance within the proposed model family and suggest that modern Hopfield networks are a promising component for EEG-based dementia classification. 

The increasing demand for personalized, non-invasive, and real-time medical monitoring has motivated the integration of digital twin technologies into healthcare systems. This paper proposes a Digi-Phy Twin, an augmented framework that couples digital and physical twins into a closed-loop, resulting in an adaptive system for medical applications. The physical twin represents the real-world anatomical and physiological structure of the head, while the digital twin incorporates physics-based modeling, data-driven analytics, and artificial intelligence to replicate underlying mechanisms and estimate internal states. Real-time data exchange between the physical and digital domains enables continuous learning, predictive analysis, and feedback-driven adaptation. The proposed architecture supports multimodal data fusion, real-time prediction, and visualization, facilitating personalized monitoring and clinical decision support. Key challenges related to privacy, interoperability, scalability, and power-efficient computation are discussed. The Digi-Phy Twin framework establishes a foundation for next-generation intelligent healthcare systems and demonstrates strong potential for non-invasive brain monitoring and precision medicine applications. 

Speech imagery (SI) has emerged as a promising paradigm for brain–computer interfaces (BCIs) that enable hands-free communication without external stimuli. Classification of electroencephalography-based SI signals (SI-EEG) remains challenging due to limited data and high inter- and intra-subject variability. Most existing approaches rely on manually designed architectures, which is a time-consuming and labor-intensive process that requires significant domain expertise. To address these challenges, we propose a novel neural architecture search (NAS) framework for SI-EEG signal classification with bi-level optimization and path-level binarization to automatically and adaptively discover optimal model architectures. Furthermore, we propose Deformable ConvNeXt block, a novel convolutional neural network-based operation that integrates transformerinspired design principles and EEG-specific priors. This block enables each neuron to adaptively control the size of the EEG signal’s receptive field. It allows the model to more effectively capture temporal variations in mental activity in inter- and intra-subject classification settings. Evaluated across three SIEEG datasets, our approach achieves classification accuracies of 68.30%, 68.49%, and 72.32%, surpassing the state of the art. These results demonstrate the effectiveness and scalability of our NAS-based framework for next-generation real-time BCI systems.

Cognitive workload plays a critical role in human performance during complex problem solving and decision making, particularly in machine-assisted environments where excessive workload can degrade efficiency and increase fatigue. This study presents a computational approach for analyzing and classifying cognitive workload from electroencephalography signals by jointly modeling behavioral, neural, and computational characteristics of cognitive processes. The proposed method integrates spectral–spatial feature extraction, functional connectivity modeling, and temporal sequence learning within a unified deep learning framework to capture complementary aspects of brain dynamics. Experimental evaluation is conducted on brain signal data collected from participants performing Arithmetic and Sudoku tasks under varying difficulty levels and assistance conditions. The results demonstrate that the proposed method achieves high classification accuracy (92.3% for Arithmetic and 90.8% for Sudoku) and consistently outperforms strong baseline models. In addition, the analysis reveals that increasing task difficulty leads to degraded performance and increased cognitive demand, while assistance helps restore performance and reduce perceived workload. These findings highlight the effectiveness of the proposed approach for cognitive workload monitoring and its potential for supporting adaptive machine-assisted systems.

Driver drowsiness is a leading cause of traffic accidents, necessitating real-time, reliable detection systems to ensure road safety. This study proposes a Modified TSception architecture for robust assessment of driver fatigue and mental workload using Electroencephalography (EEG). The model introduces a five-layer hierarchical temporal refinement strategy to capture multi-scale brain dynamics, surpassing the original TSception’s three-layer approach. Key innovations include the use of Adaptive Average Pooling (ADP) for structural flexibility across varying EEG dimensions and a two-stage fusion mechanism to optimize spatiotemporal feature integration for improved stability. Evaluated on the SEED-VIG dataset, the Modified TSception achieves 83.46% accuracy, comparable to the original model (83.15%), but with a significantly reduced confidence interval (0.24 vs. 0.36), indicating better performance stability. The architecture’s generalizability was further validated on the STEW mental workload dataset, achieving state-of-the-art accuracies of 95.93% and 95.35% for 2-class and 3-class classification, respectively. These results show that the proposed modifications improve consistency and cross-task generalizability, making the model a reliable framework for EEG-based safety monitoring.

Aims: Acute myocardial infarction (AMI) remains a leading global cause of mortality, where timely diagnosis is critical to enable early intervention. The 12-lead electrocardiogram (ECG) is a critical tool for AMI detection. While deep learning (DL) models show promise for automated ECG analysis, most prior studies rely on small, curated datasets with limited external validation, limiting their clinical applicability.

Methods and results: We developed a multimodal DL model (Conv-BiLSTM-Attn) integrating convolutional and recurrent neural networks with an attention mechanism. Using a large, multi-centre dataset of 145 656 ECGs from 96 813 patients across three Swedish hospitals. We trained the model to detect AMI using raw 12-lead ECG signals and demographic inputs (age, sex). Model performance was evaluated under two external validation protocols: generalization across hospitals (GAH) and leave-one-hospital-out (LOHO). The model achieved an area under the receiver operating characteristic (AUROC) of 0.848 (95% CI: 0.84–0.86) and an area under the precision-recall (AUPRC) of 0.456, reflecting class imbalance (∼6–10% AMI prevalence) under the GAH protocol. Subgroup AUROCs ranged from 0.79 to 0.92 across age and sex groups. At the Youden-optimized threshold (0.439), the model showed sensitivity of 0.736, specificity of 0.793, negative predictive value of 0.976, and weighted F1-score of 0.837. Under the LOHO protocol, AUROCs ranged from 0.801–0.849. At Youden thresholds (0.34–0.50), sensitivity ranged from 0.671 to 0.776 and specificity from 0.651 to 0.801, confirming generalizability across sites. Conv-BiLSTM-Attn outperformed benchmark models, with Score-CAM highlighting relevant ST–T segments.

Conclusion: This DL model can support accurate and generalizable AMI detection from routine ECGs, with the Conv-BiLSTM-Attn architecture outperforming current benchmark approaches.

Integration of multiple technologies amid Industry 5.0 is observed in most of the engineering application. However, Machine Learning (ML) and Digital Twin Technology (DTT) play a major role in data acquisition, processing, data analytics and decision making process of almost every automated system. Due to the gigantic amount of data generated in healthcare and medicine applications, manual analysis of this data becomes very cumbersome and time consuming. Therefore, ML, artificial intelligence (AI), data science, DTT and Cyber Physical Systems (CPSs) emerge as powerful tools for efficient, accurate, automated and fast processing of this medical data including 1D, 2D signals as well as multidimensional images. This paper presents an overview and comparative analysis of machine learning algorithms and tools utilized for brain monitoring applications especially focusing on microwave techniques. A systematic review and meta analysis based on PRISMA approach is presented and analyzed. A brief outline of ML algorithms utilized for different applications focusing on brain temperature measurement, stroke detection, and Intracranial Pressure (ICP) measurement are elaborated along with ML tools for microwave imaging. 

Inner speech (IS), or imagined speech without overt articulation, is a promising target for brain-computer interfaces (BCIs) aimed at restoring communication in individuals with severe speech impairments, such as locked-in syndrome. Foundation models (FMs), typically trained using self-supervised learning (SSL) on large-scale datasets, offer new opportunities for learning transferable and robust representations from neural signals. This mini review provides an overview of FM-based approaches for IS decoding using non-invasive neuroimaging modalities, including functional magnetic resonance imaging, electroencephalography, magnetoencephalography, and functional near-infrared spectroscopy, highlighting architectural trends, pretraining strategies, and model adaptation techniques. We discuss how recent models move beyond task-specific classification toward scalable representation learning and semantic-level decoding. Despite these advances, several challenges remain, including the weak, noisy, and non-stationary nature of neural signals, variability in data acquisition, and limitations in dataset scale, standardization, computational resources, interpretability, and evaluation metrics. Ethical and privacy considerations are also critical. Overall, FMs provide a promising paradigm for non-invasive IS decoding, addressing neurophysiological, methodological, and ethical challenges is essential for developing scalable and reliable BCI systems.

Existing self-supervised adversarial training (self-AT) methods rely on hand-crafted adversarial attack strategies for PGD attacks, which fail to adapt to the evolving learning dynamics of the model and do not account for instance-specific characteristics of images. This results in sub-optimal adversarial robustness and limits the alignment between clean and adversarial data distributions. To address this, we propose ASTrA (Adversarial Self-supervised Training with Adaptive-Attacks), a novel framework introducing a learnable, self-supervised attack strategy network that autonomously discovers optimal attack parameters through exploration-exploitation in a single training episode. ASTrA leverages a reward mechanism based on contrastive loss, optimized with REINFORCE, enabling adaptive attack strategies without labeled data or additional hyperparameters. We further introduce a mixed contrastive objective to align the distribution of clean and adversarial examples in representation space. ASTrA achieves state-of-the-art results on CIFAR10, CIFAR100, and STL10 while integrating seamlessly as a plug-and-play module for other self-AT methods. ASTrA shows scalability to larger datasets, demonstrates strong semi-supervised performance, and is resilient to robust overfitting, backed by explainability analysis on optimal attack strategies. Project page for source code and other details at https://prakashchhipa.github.io/projects/ASTrA.

Children diagnosed with Attention-Deficit/Hyperactivity Disorder (ADHD) face many difficulties in maintaining their concentration (in terms of attention levels) and controlling their behaviors. Previous studies have mainly focused on identifying brain regions involved in cognitive processes or classifying ADHD and control subjects. However, the classification of attention levels of ADHD subjects has not yet been explored. Here, a robust Swin Transformer (Swin-T) model is proposed to classify the attention levels of ADHD subjects. The experimental cognitive task ‘Surround suppression’ includes two events: Stim ON and Stim OFF related to the high and low attention levels of a subject. In the proposed framework, ADHD-specific channels are initially identified from input Electroencephalography (EEG). Next, the significant, non-noisy connectivity features are extracted from those channels through the Singular Value Decomposition (SVD) method. Finally, the non-noisy features are passed to the robust Swin-T model for attention-level classification. The proposed model achieves 97.28% classification accuracy with 12 subjects. The robustness of the proposed model leads to potential benefits in EEG-based research and clinical settings, enhancing the reliability of ADHD assessments.