Search Results (92)

Copy number variants (CNVs) are genomic rearrangements that carry a substantial risk for neurodevelopmental and neuropsychiatric disorders. Among these, recurrent deletions and duplications at the 16p11.2 locus are robustly associated with autism spectrum disorders, schizophrenia, epilepsy, and related conditions, yet also display marked variability in penetrance and phenotypic expression. Accumulating evidence indicates that 16p11.2 gene dosage influences multiple stages of brain development, from early progenitor dynamics and neuronal migration to synaptic formation, refinement, and plasticity. However, how disruptions across these processes are integrated over time, and how they relate to the observed variability and incomplete penetrance, remains poorly understood. In this review, we summarize the current evidence on the impact of 16p11.2 CNVs on brain development, focusing on cellular and circuit-level processes that shape neural connectivity. We discuss how gene dosage imbalance influences early developmental trajectories, synaptic formation and pruning, interneuron maturation, and activity-dependent plasticity, and consider how these processes interact across developmental stages. We suggest a conceptual framework wherein 16p11.2 CNVs do not impose fixed pathogenic outcomes, but rather they contribute towards developmental constraints that shape the timing and stability of neural circuit development. Consequently, these constraints increase vulnerability to neurodevelopmental and psychiatric outcomes in a context-dependent manner. Full article

As disaster imagery and social media reports become vital for crisis response, automated assessment systems must address challenges of multimodal integration, privacy-aware learning, and computational efficiency. To address these challenges, we propose TriDA, a privacy-aware and efficiency-conscious multimodal disaster classification framework that fuses image features with text representations through a late-fusion design. A classifier-head DP-SGD stage is used to report training-record-level differential privacy accounting for paired image–text samples under the stated private optimization protocol. To study efficiency-oriented simplification, structured neuron pruning reduces redundant capacity in the classification head while preserving predictive utility. Experiments on the multimodal damage identification dataset show that TriDA maintains strong classification performance, exhibits a controlled privacy–utility trade-off under increasing DP noise, and achieves quantifiable classifier-head parameter and MAC reductions through pruning. These findings position TriDA as a controlled empirical framework for privacy-aware and resource-conscious multimodal disaster assessment. Full article

A novel double-criterion early stopping pruning strategy is introduced in this study. The proposed approach enables the structural optimization of a hybrid filtering framework that integrates FeedForward Neural Networks with an adaptive Extended Kalman Filter, by simultaneously monitoring the validation error and the trace of the error covariance matrix. Unlike classical pruning methods, which are applied after the completion of the training process and aggressively remove network neurons, the proposed scheme exploits the learning procedure, achieving a more selective reduction of 2% to 13%, balancing effectively between strong generalization performance and computationally efficient training. The proposed framework is evaluated on wind speed forecasts obtained from a numerical weather prediction model, within a time-varying window scheme, demonstrating promising improvements. Key statistical indices, such as the Mean Absolute Error and the Root Mean Square Error, were significantly reduced, with reductions ranging from approximately 65% to 80% and 60% to 78%, respectively. These findings suggest that the proposed methodology offers a robust and accurate framework for time series forecasting in operational settings. Full article

Land cover classification (LCC) is a fundamental task in remote sensing, which enables effective environmental monitoring, agricultural planning, and disaster management. The existing approaches often rely on fine-tuning pre-trained models, which are not specifically designed for LCC, which lead to suboptimal performance in complex scenarios. To address these limitations, we propose the Cross-Module Attention Land Cover Network (CALCNet), a novel architecture developed from scratch. CALCNet follows a contracting and restoration backbone, where the contracting path extracts progressively abstract semantic features while reducing spatial resolution, and the restoration path recovers fine-grained spatial details through upsampling and skip connections. In addition, CALCNet integrates a cross-module attention mechanism that combines spatial attention and multi-scale feature selection to enhance feature representation. Furthermore, we applied a differential evolution-based neuron pruning strategy to create a compressed CALCNet variant, which retains high classification performance while reducing computational cost. The CALCNet is evaluated on four benchmark LCC datasets, AID, UCMerced_LandUse, NWPU_RESISC45, and EuroSAT, demonstrating strong performance across all benchmarks. Specifically, the model achieves classification accuracies of 98.09%, 99.47%, 99.19%, and 99.19%, respectively. The compressed CALCNet variant reduces computational cost to 78.55 million floating point operations (FLOPs) with a model size of 43 MB, while achieving improved inference speeds (38.32 frames/sec on CPU and 118.3 frames/sec on GPU), representing approximately 45–50% reduction in FLOPs and model storage. These results highlight that CALCNet is both highly accurate and computationally efficient, making it well suited for real-world LCC applications. Full article

Accurate and computationally efficient trajectory tracking remains a critical challenge for unmanned aerial vehicles (UAVs), particularly when nonlinear model predictive control (NMPC) is combined with learning-based dynamics models that introduce significant computational burden. This paper proposes a sparse neural dynamics modeling approach by integrating structured pruning and robustness-enhancing fine-tuning techniques to enable efficient nonlinear MPC (NMPC) for UAV trajectory tracking. To this end, a structured neuron-level pruning strategy is introduced, combining L1-norm importance scores with adversarial sensitivity analysis to identify and remove redundant neurons from a neural dynamics model. To preserve smoothness and robustness in closed-loop control, spectral norm constraints and gradient regularization are further incorporated during fine-tuning. The resulting pruned neural dynamics model is embedded into an NMPC framework for online trajectory tracking. Simulation results on a fixed-wing UAV demonstrate that the proposed method reduces the number of trainable parameters by approximately 69% and achieves a 19% reduction in average NMPC solve time, leading to an effective control update frequency of about 39 Hz under the considered simulation settings. Compared with conventional controllers, including TECS and linear MPC, the proposed approach achieves significantly improved trajectory tracking accuracy, as reflected by lower MAE and RMSE across all position axes. These results indicate that structured sparsification of neural dynamics models provides an effective means to enhance both computational efficiency and tracking performance in NMPC-based UAV control. Full article

Spiking Neural Networks (SNNs) have attracted increasing attention due to their energy efficiency and suitability for neuromorphic data processing. Despite these advantages, the security of SNNs—particularly their robustness against backdoor attacks—remains underexplored. This study revisits fine-pruning, a widely adopted backdoor defense technique in deep neural networks, and adapts it to the unique spatio-temporal characteristics of SNNs. We propose two SNN-specific fine-pruning methods: Hook–Surrogate Gradient-based fine-pruning (HS-FP) and Spike–STDP-based fine-pruning (SS-FP). HS-FP leverages hook-based activation analysis with surrogate gradient learning, while SS-FP integrates total spike activity with hybrid STDP and surrogate gradient fine-tuning. We evaluate both methods against static, moving, and smart backdoor attacks on two neuromorphic benchmarks, N-MNIST and DVS128-Gesture. Experimental results show that both approaches reduce the attack success rate down to approximately 10% while preserving model accuracy above 99% on N-MNIST and achieving substantial recovery on DVS128-Gesture. Moreover, our analysis reveals that several phenomena observed in fine-pruning-based defenses for deep neural networks—such as mixed-function neurons and backdoor reactivation during fine-tuning—also manifest in SNNs. These findings highlight both the effectiveness and limitations of fine-pruning in the SNN domain and suggest promising directions for extending existing DNN security methodologies to neuromorphic systems. Full article

Alzheimer’s disease (AD) is traditionally defined by Amyloid-β (Aβ) plaques and tau neurofibrillary tangles, yet these proteinopathies alone fail to explain disease heterogeneity, progression, and cognitive decline. Emerging evidence identifies chronic neuroinflammation as a central integrator that converts molecular pathology into synaptic failure and neurodegeneration. In this context, Aβ acts as a danger-associated molecular pattern that activates microglial and astrocytic immune programs through receptors such as TREM2, TLRs, and RAGE, leading to inflammasome activation, cytokine release, and oxidative stress. These responses pathologically re-engage developmental complement pathways (C1q–C3–CR3), driving excessive synaptic pruning that correlates more closely with cognitive impairment than neuronal loss. Reactive astrocytes further amplify dysfunction by impairing glutamate and potassium homeostasis, promoting excitotoxic and metabolic stress, while inflammatory glia facilitate prion-like tau propagation via extracellular vesicles. Concurrent neurovascular inflammation disrupts blood–brain barrier integrity and cerebral perfusion, reinforcing immune-metabolic failure. Importantly, neuroinflammatory biomarkers (GFAP, sTREM2, YKL-40, cytokines, complement, and TSPO-PET) provide dynamic readouts of disease activity and therapeutic response. Together, these findings position AD as a disorder of failed immune resolution and support precision immunomodulatory and pro-resolving therapies aimed at restoring neuroimmune homeostasis rather than merely removing protein aggregates. Full article

Neuronal synapses are the functional units of communication in the central nervous system. This review describes the molecular mechanisms regulating synaptic transmission, plasticity, and circuit refinement. At the presynaptic active zone, scaffolding proteins including bassoon, piccolo, RIMs, and munc13 organize vesicle priming and the localization of voltage-gated calcium channels. Neurotransmitter release is mediated by the SNARE complex, comprising syntaxin-1, SNAP25, and synaptobrevin, and triggered by the calcium sensor synaptotagmin-1. Following exocytosis, synaptic vesicles are recovered through clathrin-mediated, ultrafast, bulk, or kiss-and-run endocytic pathways. Postsynaptically, the postsynaptic density (PSD) serves as a protein hub where scaffolds such as PSD-95, shank, homer, and gephyrin anchor excitatory (AMPA, NMDA) and inhibitory (GABA-A, Glycine) receptors are observed. Synaptic strength is modified during long-term potentiation (LTP) and depression (LTD) through signaling cascades involving kinases like CaMKII, PKA, and PKC, or phosphatases such as PP1 and calcineurin. These pathways regulate receptor trafficking, Arc-mediated endocytosis, and actin-dependent remodeling of dendritic spines. Additionally, synapse formation and elimination are guided by cell adhesion molecules, including neurexins and neuroligins, and by microglial pruning via the complement cascade (C1q, C3) and “don’t eat me” signals like CD47. Molecular diversity is further expanded by alternative splicing and post-translational modifications. A unified model of synaptic homeostasis is required to understand the basis of neuropsychiatric and neurological disorders. Full article

Bidirectional Encoder Representations from Transformers (BERT) excels in natural language processing (NLP) but is costly on edge devices. We introduce an unsupervised, retraining-free structured pruning scheme for BERT, guided by mutual information (MI). Leveraging Rényi $\alpha$-order entropy, we design a representation-aware MI estimator and a principled kernel-bandwidth selection, producing stable, sample-efficient neuron-level pruning signals. This method removes redundant units while preserving representational capacity, reduces memory and latency, and deploys readily on commodity hardware. Explainable-AI visualizations clarify how compression reshapes intermediate features and predictions. Across benchmarks, the compressed models maintain minimal accuracy loss, outperform or match strong unsupervised baselines, and remain competitive with supervised alternatives. Full article

Deep neural networks (DNNs) are highly effective across many domains but are sensitive to noisy or corrupted training data. Existing noise mitigation strategies often rely on strong assumptions about noise distributions or require costly retraining, limiting their scalability. Inspired by machine unlearning, we propose a novel framework that integrates attribution-guided data partitioning, neuron pruning, and targeted fine-tuning to enhance robustness. Our method uses gradient-based attribution to probabilistically identify clean samples without assuming specific noise characteristics. It then applies sensitivity-based neuron pruning to remove components most susceptible to noise, followed by fine-tuning on the retained high-quality subset. This approach jointly addresses data and model-level noise, offering a practical alternative to full retraining or explicit noise modeling. We evaluate our method on CIFAR-10 image classification and keyword spotting tasks under varying levels of label corruption. On CIFAR-10, our framework improves accuracy by up to 10% (F-FT vs. retrain) and reduces retraining time by 47% (L-FT vs. retrain), highlighting both accuracy and efficiency gains. These results highlight its effectiveness and efficiency in noisy settings, making it a scalable solution for robust generalization. Full article

Neurodegenerative diseases, including Alzheimer’s disease (AD), Parkinson’s disease (PD), and amyotrophic lateral sclerosis (ALS), represent significant global health challenges, affecting millions and straining healthcare systems. These disorders involve progressive neuronal loss and cognitive decline, with incompletely elucidated underlying mechanisms. Chronic neuroinflammation is increasingly recognized as a critical contributor to disease progression. The brain’s resident immune cells, microglia, are central to this inflammatory response. When overactivated, microglia and other immune cells, such as peripheral macrophages, can exacerbate inflammation and accelerate disease development. Triggering Receptor Expressed on Myeloid Cells 2 (TREM2) is a transmembrane receptor of the immunoglobulin superfamily that demonstrates high expression on microglia in the central nervous system. TREM2 serves a vital role in regulating phagocytosis, synaptic pruning, and energy metabolism. This review examines the functions of TREM2 in neurodegenerative diseases and its potential as a therapeutic target, aiming to inform future treatment strategies. Full article

Microglia and macrophages are critical immune cells within the central nervous system (CNS), with distinct roles in development, homeostasis, and disease. Once viewed as passive bystanders, these cells are now recognized for their dynamic phenotypic plasticity, which enables them to respond to a wide range of physiological and pathological stimuli. During homeostasis, microglia and CNS-resident macrophages actively participate in synaptic pruning, neuronal support, myelin regulation, and immune surveillance, contributing to CNS integrity. However, under pathological conditions, these cells can adopt neurotoxic phenotypes, exacerbating neuroinflammation, oxidative stress, and neuronal damage in diseases such as Alzheimer’s, Parkinson’s, multiple sclerosis, and glioblastoma. This review synthesizes emerging insights into the molecular, epigenetic, and metabolic mechanisms that govern the behavior of microglia and macrophages, highlighting their developmental origins, niche-specific programming, and interactions with other CNS cells. We also explore novel therapeutic strategies aimed at modulating these immune cells to restore CNS homeostasis, including nanotechnology-based approaches for selective targeting, reprogramming, and imaging. Understanding the complex roles of microglia and macrophages in both health and disease is crucial for the development of precise therapies targeting neuroimmune interfaces. Continued advances in single-cell technologies and nanomedicine are paving the way for future therapeutic interventions in neurological disorders. Full article

Neurodegeneration is increasingly recognized not as a linear trajectory of protein accumulation, but as a multidimensional collapse of biological organization—spanning intracellular signaling, transcriptional identity, proteostatic integrity, organelle communication, and network-level computation. This review intends to synthesize emerging frameworks that reposition neurodegenerative diseases (ND) as progressive breakdowns of interpretive cellular logic, rather than mere terminal consequences of protein aggregation or synaptic attrition. The discussion aims to provide a detailed mapping of how critical signaling pathways—including PI3K–AKT–mTOR, MAPK, Wnt/β-catenin, and integrated stress response cascades—undergo spatial and temporal disintegration. Special attention is directed toward the roles of RNA-binding proteins (e.g., TDP-43, FUS, ELAVL2), m6A epitranscriptomic modifiers (METTL3, YTHDF1, IGF2BP1), and non-canonical post-translational modifications (SUMOylation, crotonylation) in disrupting translation fidelity, proteostasis, and subcellular targeting. At the organelle level, the review seeks to highlight how the failure of ribosome-associated quality control (RQC), autophagosome–lysosome fusion machinery (STX17, SNAP29), and mitochondrial import/export systems (TIM/TOM complexes) generates cumulative stress and impairs neuronal triage. These dysfunctions are compounded by mitochondrial protease overload (LONP1, CLPP), UPR maladaptation, and phase-transitioned stress granules that sequester nucleocytoplasmic transport proteins and ribosomal subunits, especially in ALS and FTD contexts. Synaptic disassembly is treated not only as a downstream event, but as an early tipping point, driven by impaired PSD scaffolding, aberrant endosomal recycling (Rab5, Rab11), complement-mediated pruning (C1q/C3–CR3 axis), and excitatory–inhibitory imbalance linked to parvalbumin interneuron decay. Using insights from single-cell and spatial transcriptomics, the review illustrates how regional vulnerability to proteostatic and metabolic stress converges with signaling noise to produce entropic attractor collapse within core networks such as the DMN, SN, and FPCN. By framing neurodegeneration as an active loss of cellular and network “meaning-making”—a collapse of coordinated signal interpretation, triage prioritization, and adaptive response—the review aims to support a more integrative conceptual model. In this context, therapeutic direction may shift from damage containment toward restoring high-dimensional neuronal agency, via strategies that include the following elements: reprogrammable proteome-targeting agents (e.g., PROTACs), engineered autophagy adaptors, CRISPR-based BDNF enhancers, mitochondrial gatekeeping stabilizers, and glial-exosome neuroengineering. This synthesis intends to offer a translational scaffold for viewing neurodegeneration as not only a disorder of accumulation but as a systems-level failure of cellular reasoning—a perspective that may inform future efforts in resilience-based intervention and precision neurorestoration. Full article

We explore the sparsity and prunability of multi-layer perceptrons (MLPs) trained using the Forward-Forward (FF) algorithm, an alternative to backpropagation (BP) that replaces the backward pass with local, contrastive updates at each layer. We analyze the sparsity of the weight matrices during training using multiple metrics, and test the prunability of FF networks on the MNIST, FashionMNIST and CIFAR-10 datasets. We also propose FFLib—a novel, modular PyTorch-based library for developing, training and analyzing FF models along with a suite of FF-based architectures, including FFNN, FFNN+C and FFRNN. In addition to structural sparsity, we describe and apply a new method for visualizing the functional sparsity of neural activations across different architectures using the HSV color space. Moreover, we conduct a sensitivity analysis to assess the impact of hyperparameters on model performance and sparsity. Finally, we perform pruning experiments, showing that simple FF-based MLPs exhibit significantly greater robustness to one-shot neuron pruning than traditional BP-trained networks, and a possible 8-fold increase in compression ratios while maintaining comparable accuracy on the MNIST dataset. Full article

The complement system is a vital component of innate immunity. Besides its roles in pathogen defense, its significance in neurodevelopment, neurodegeneration, and cancer progression is beginning to be recognized. We performed a comprehensive literature review to summarize the involvement and dysregulation of the complement system in three main CNS-associated conditions: Alzheimer’s disease, schizophrenia, and glioma. In Alzheimer’s disease, activation of the complement system contributes to neuroinflammation, synaptic loss, and neuronal death. In glioblastoma, complement promotes tumor growth, immune evasion, and therapy resistance. In schizophrenia, genetic variations in complement components, particularly C4A, are associated with synaptic pruning abnormalities and disease susceptibility. We conclude that the complement system has a dual role of protector and pathogenic mediator in the central nervous system. While it is critical in neurodegenerative, oncological, and psychiatric disorders, its role is not understood well enough. For therapeutic purposes, targeting the complement system may open new frontiers for therapeutic interventions without disrupting important physiological processes. More research is needed to elucidate the exact roles of the complement and help translate these findings into clinical settings. Full article