Search Results (2,447)

Vertebral compression fractures (VCFs) are one of the most common spinal disorders encountered clinically. Untimely diagnosis or inaccurate classification often leads to prolonged pain and functional impairment in patients. To enhance diagnostic accuracy and efficiency, this study addressed the high cost and limited applicability of computed tomography (CT) and magnetic resonance imaging (MRI) examinations by leveraging the universality and convenience of X-ray imaging. We proposed a multi-stage deep learning-based method for identifying vertebral compression fractures. The method first employs Discrete Wavelet Transform-YOLOv5 (DWT-YOLOv5) for preliminary vertebral region localization, followed by Polarized Self-Attention-UNet (PSA-UNet) for precise segmentation. Finally, a ResNet50 network incorporating a Convolutional Block Attention Module (CBAM) performs graded classification, categorizing vertebrae into four types: Non-fracture, Mild fracture, Moderate fracture, and Severe fracture. The experimental results demonstrate that the proposed method achieved average accuracy, precision, recall, specificity, and F1-score of 83.7%, 88.1%, 86.2%, 97.7%, and 87.2%, respectively. The proposed method fully leverages the cost-effectiveness and convenience of X-ray imaging, providing clinicians with an efficient and economical auxiliary diagnostic tool. It enables rapid and accurate identification of vertebral compression fractures in emergency and initial screening scenarios. Full article

Background/Objectives: Insomnia is one of the most prevalent and persistent symptoms among patients with breast cancer, yet it remains under-recognized and undertreated in routine clinical practice. Beyond its impact on sleep quality, insomnia is increasingly understood as a multidimensional condition involving neurobiological, psychological, and behavioral mechanisms, closely intertwined with cancer-related stress and psychiatric comorbidities. This narrative review aims to provide a comprehensive and integrative overview of insomnia in breast cancer, focusing on its epidemiology, pathophysiological underpinnings, neuropsychiatric correlates, and clinical implications, while highlighting gaps in current research and management. Methods: A narrative review of the literature was conducted, including studies published in major medical databases (PubMed, Scopus, and Web of Science) up to 2025. Relevant articles addressing insomnia, sleep disturbances, psychiatric symptoms, and neurobiological mechanisms in breast cancer populations were selected and synthesized. Results: Insomnia affects a substantial proportion of breast cancer patients across the disease trajectory, from diagnosis to survivorship. Its etiology is multifactorial, involving dysregulation of the hypothalamic–pituitary–adrenal axis, inflammatory processes, and circadian rhythm, as well as treatment-related factors such as chemotherapy, endocrine therapy, and menopausal symptoms. Insomnia frequently co-occurs with depression, anxiety, fatigue, and pain, forming symptom clusters that significantly impair quality of life and may influence clinical outcomes. Emerging evidence supports a bidirectional relationship between insomnia and psychiatric vulnerability, suggesting a shared neurobiological substrate within the brain–body stress axis. Conclusions: Insomnia in breast cancer should be conceptualized as a neuropsychiatric condition embedded within a broader stress-related symptom network rather than as an isolated sleep disturbance. Improved screening, interdisciplinary management, and the integration of evidence-based interventions such as cognitive behavioral therapy for insomnia are essential. Research should focus on personalized and mechanistically informed approaches to better address this highly prevalent yet insufficiently managed condition. Full article

Automatic Modulation Classification (AMC) is essential for waveform-level signal characterization. It supports spectrum sensing, signal identification, and adaptive resource allocation in cognitive radio and next-generation wireless systems. However, channel impairments such as multipath propagation, frequency offset, fast fading, and noise degrade modulation signatures, making reliable AMC challenging. Existing deep learning-based approaches often rely on purely data-driven learning, leading to insufficient modeling of modulation-relevant features, loss of transient characteristics, and limited exploitation of hierarchical relationships among modulation types. To address these issues, this paper proposes PHM-Net, a physics-informed hierarchical multi-scale network for robust AMC. The model employs a hierarchical backbone with residual encoder blocks. A Transient Feature Gating (TFG) module enhances modulation-relevant representations, a Cross-Resolution Signal Aggregation (CRSA) module fuses multi-stage features, and a Physics-Informed Hierarchical Loss (PI-HL) enforces consistency between coarse- and fine-grained predictions. Experimental results on three benchmark datasets (RML2016.10a, RML2016.10b, and RML2018.01a) show that PHM-Net consistently achieves the highest average accuracy among all compared models. On RML2018.01a, which contains 1024-sample sequences and 24 classes, PHM-Net achieves an average accuracy of 64.59% and a best-case accuracy of 98.42%, surpassing AMC_Net by 11.14 and 17.09 percentage points and CNN-Transformer by 9.43 and 11.15 percentage points, respectively. PHM-Net provides a robust and interpretable solution for AMC under complex channel conditions. Full article

Obesity is a major risk factor for colorectal cancer (CRC) and is increasingly recognized as a contributor to cancer-related cognitive impairment; however, the mechanistic pathways linking metabolic dysfunction, tumor progression, and brain dysfunction remain incompletely defined. Emerging evidence indicates that obesity-induced gut microbial dysbiosis and intestinal barrier disruption may serve as a biologically plausible mechanism connecting these processes via the gut–brain axis although direct clinical causality remains to be firmly established. In obesity, alterations in gut microbiota composition characterized by depletion of barrier-protective taxa and enrichment of pro-inflammatory and genotoxic pathobionts compromise epithelial tight-junction integrity and promote metabolic endotoxemia. The translocation of microbial products, including lipopolysaccharide, sustains chronic systemic inflammation, accelerates CRC progression, and remodels the tumor microenvironment. Notably, these peripheral inflammatory signals extend beyond the intestine and tumor, disrupting blood–brain barrier integrity, activating microglia and astrocytes, and impairing synaptic plasticity within hippocampal and frontal networks. Clinically, these processes manifest as cancer-related cognitive impairment (CRCI), with predominant deficits in attention, processing speed, and working memory, which are often detectable around the time of diagnosis and independent of chemotherapy exposure. This review synthesizes in vivo, in vitro, and human evidence into a proposed theoretical “two-barrier failure” model of obesity-associated CRC and cognitive dysfunction. In addition to mechanistic synthesis, we discuss barrier-centered therapeutic strategies, including targeted probiotics, postbiotics, SCFA supplementation, obesity management through dietary and weight-loss interventions, and potential pharmacological approaches to epithelial and neurovascular barrier protection. We also outline testable clinical trial designs for evaluating these interventions in obesity-associated CRC. Full article

Chronic kidney disease (CKD) affects approximately 700 million people worldwide and is a major contributor to end-stage renal disease (ESRD), cardiovascular morbidity, and premature mortality. Although oxidative stress has long been considered central to CKD progression, conventional antioxidant strategies have not consistently improved clinical outcomes, suggesting that excess reactive oxygen species (ROS) alone cannot fully account for the underlying disease pathophysiology. Emerging evidence supports a broader paradigm of redox network failure, characterized by the disruption of coordinated signaling among ROS, nitric oxide (NO), and reactive sulfur species (RSS). Within this framework, hydrogen sulfide (H₂S), a major endogenous RSS, functions as a key regulator of renal redox homeostasis. CKD is consistently associated with systemic and renal H₂S deficiency, accompanied by downregulation of cystathionine β-synthase (CBS), cystathionine γ-lyase (CSE), and 3-mercaptopyruvate sulfurtransferase (3-MST), as well as impaired transsulfuration and disrupted mitochondrial sulfide oxidation. Importantly, this deficiency cannot be explained solely by reduced renal function but instead reflects active suppression of H₂S biosynthesis. Uremic toxins, particularly indoxyl sulfate (IS), contribute to this process through activation of the aryl hydrocarbon receptor (AhR), which inhibits specificity protein 1 (Sp1)-dependent transcription of H₂S-producing enzymes. This IS–AhR–Sp1 axis provides a mechanistic link between toxin accumulation and disruption of the sulfur arm of the redox network, amplifying oxidative stress, endothelial dysfunction, mitochondrial impairment, ferroptotic vulnerability, and fibrotic remodeling. Beyond H₂S itself, downstream RSS, including persulfides, polysulfides, and thiosulfate, may represent the principal bioactive mediators of sulfur-dependent redox signaling, and their coordinated depletion in CKD may impair redox buffering capacity beyond what H₂S measurement alone reflects. This review integrates current evidence to propose a conceptual model in which CKD progression involves failure of coordinated redox signaling—characterized by feed-forward network collapse and threshold-dependent transition to a self-sustaining high-ROS state—with H₂S deficiency representing one mechanistically supported component of this broader network disruption. This framework highlights the therapeutic potential of targeting redox network restoration rather than isolated oxidative pathways in CKD. Full article

Post-COVID-19 syndrome (PCS), also referred to as post-acute sequelae of SARS-CoV-2 infection (PASC), represents a heterogeneous set of persistent clinical manifestations developing after acute infection. These conditions are associated with immune dysregulation, autonomic imbalance, impaired thymic function, and possible viral persistence. Objective: This study aims to systematically synthesise current evidence on the immunopathogenesis of PCS and to critically evaluate the application of artificial intelligence (AI) and machine learning (ML) approaches for its prediction and clinical stratification. Methods: A PRISMA 2020–informed systematic review was conducted using PubMed/MEDLINE, Scopus, Web of Science, elibrary.ru and Embase databases (January 2020–December 2025). Studies addressing immunopathological mechanisms and AI/ML applications in PCS were selected based on predefined eligibility criteria. Risk of bias in prediction studies was assessed using the PROBAST tool. Due to heterogeneity, a structured qualitative synthesis was performed. Current evidence indicates that PCS may result from sustained systemic inflammation, cytokine dysregulation, autoimmunity, and delayed restoration of T-cell homeostasis, including reduced thymic output of naïve T lymphocytes. Persistent thymic dysfunction may contribute to prolonged immune imbalance, increased susceptibility to secondary infections, and reactivation of latent viruses. AI/ML approaches—including gradient boosting, ensemble learning, deep neural networks, and natural language processing—have demonstrated promising performance across multimodal datasets. However, significant limitations were identified, including small sample sizes, overfitting, lack of external validation, and heterogeneity in outcome definitions. Conclusions: The integration of immunopathological insights with data-driven modelling highlights the potential of combined approaches for improving PCS risk stratification. However, current AI models remain insufficiently validated for clinical implementation. Future research should prioritise methodological standardisation, external validation, and incorporation of mechanistically informed biomarkers. Full article

Network Music Performance (NMP) refers to remote musical collaboration over a network in applications such as music education, music production, and live performance. In NMP, synchronisation is a critical factor in musicians’ Quality of Experience (QoE). This interpersonal coordination of musical actions is highly sensitive to variable network conditions, particularly to end-to-end delay and signal degradation. Existing evaluations rely mainly on subjective questionnaires or isolated objective descriptors, creating a gap for a unified metric that quantifies synchrony directly from performance signals. To address this gap, we propose the Objective Synchrony Index (OSI), an AI-driven metric that quantifies ensemble synchrony from paired NMP recordings. We computed OSI using a two-tower multi-task convolutional recurrent neural network (CRNN) that estimates synchrony-relevant descriptors from paired Musician A and Musician B audio streams. We introduce two OSI variants: timing-OSI, which captures temporal coordination through offsets, onsets, beats, and tempo coherence; and ensemble-OSI, which extends this formulation by integrating chord agreement and signal fidelity to reflect structural and perceptual aspects of ensemble interaction. We evaluated OSI using recordings from two NMP studies in which eleven pairs of musicians performed under systematically varied delay and sampling-rate conditions. After each performance, musicians completed QoE questionnaires, allowing us to relate OSI and its components to subjective ratings using repeated-measures correlation. Results showed that, under delay, timing-OSI decreases as latency increases and demonstrates construct validity against subjective QoE measures. Higher synchrony-OSI was associated with greater perceived synchronisation and satisfaction, and with lower perceived delay, irritation, and effort to follow a partner. These relationships were most consistent for offset synchrony and most selective for onset synchrony, while beat and tempo remained relatively stable. Under audio-quality degradation, ensemble-OSI remained relatively stable across sampling rates and did not significantly track subjective QoE as a single predictor. Instead, modest component-level associations suggested that satisfaction was higher when temporal stability and fidelity were preserved, whereas irritation was more closely related to reduced chord agreement. Together, these findings support timing-OSI as a promising objective synchrony metric for delay-impaired NMP, while showing that the extended ensemble-OSI requires further perceptual calibration for audio-quality degradations. Full article

Alzheimer’s disease (AD) and type 2 diabetes mellitus (T2DM) are major global health burdens that share interconnected pathological mechanisms involving impaired insulin signaling, metabolic stress, and chronic neuroinflammation. This study applied an integrative systems biology and atomistic simulation framework to investigate bioactive compounds from Caesalpinia sappan L. targeting shared molecular regulators linking AD and T2DM. Network topology analysis identified four central hub genes, STAT3, SRC, HSP90AA1, and TP53, representing key regulatory nodes involved in inflammatory signaling, kinase regulation, proteostasis, and cellular stress responses. Compound-specific interaction analysis revealed distinct target preferences among phytochemical subclasses. Protosappanin B showed strong binding toward both STAT3 and HSP90α, whereas flavonols including quercetin and rhamnetin exhibited high affinity for SRC, and the chalcone derivative sappanchalcone preferentially interacted with TP53. Atomistic molecular dynamics simulations and MM-PBSA calculations supported stable protein ligand interactions and favorable binding energetics, while density functional theory analysis indicated electronic properties consistent with sustained intermolecular interactions. Collectively, these findings suggest that structurally distinct subclasses of C. sappan phytochemicals converge on complementary regulatory hubs within the shared AD and T2DM molecular network, supporting coordinated multi-target modulation of interconnected inflammatory, kinase signaling, proteostasis, and cellular stress pathways underlying AD–T2DM comorbidity. Full article

Objective: Cerebral ischemia/reperfusion (I/R) injury triggers oxidative stress, neuroinflammation, neuronal degeneration, and white matter damage not only in directly affected cerebral regions but also in remote brain areas such as the cerebellum. Skimmianine, a naturally occurring furoquinoline alkaloid, has been reported to possess antioxidant and anti-inflammatory properties. This study investigated the protective effects of skimmianine pretreatment against secondary cerebellar injury following experimental cerebral I/R. Materials and Methods: Thirty-two female Wistar rats were randomly assigned to sham, Skimmianine, I/R, and I/R + Skimmianine groups (n = 8/group). Cerebral I/R was induced by transient middle cerebral artery occlusion for 60 min followed by 23 h reperfusion. Skimmianine (40 mg/kg/day, intraperitoneally) was administered for 14 days before ischemia induction. Oxidative stress markers, neuroinflammatory mediators, histopathological alterations, behavioral outcomes, and ultrastructural changes were evaluated. In addition, network pharmacology and molecular docking analyses were performed to explore potential molecular mechanisms. Results: Cerebral I/R significantly decreased TAS levels compared with sham (0.89 ± 0.15 vs. 1.52 ± 0.18 mmol Trolox Eq/L) and increased TOS (15.60 ± 3.03 vs. 6.80 ± 1.41 µmol H₂O₂ Eq/L), OSI (17.48 ± 0.50 vs. 4.43 ± 0.47), TNF-α (68.4 ± 10.2 vs. 18.6 ± 4.4 pg/mL), Iba1 (41.3 ± 9.7 vs. 11.7 ± 1.6 pg/mL), and GFAP levels (334.5 ± 12.5 vs. 87.7 ± 9.5 ng/mL; all p < 0.001). I/R also impaired motor performance, as shown by increased beam crossing time (11.7 ± 2.2 vs. 4.8 ± 0.7 s) and grid foot fault rate (18.6 ± 4.0% vs. 3.4 ± 1.1%). Skimmianine pretreatment significantly improved these alterations, increasing TAS to 1.29 ± 0.20 mmol Trolox Eq/L and reducing TOS, OSI, TNF-α, Iba1, and GFAP levels to 9.20 ± 2.04, 7.07 ± 0.47, 34.9 ± 7.4, 24.2 ± 6.9, and 237.0 ± 7.9, respectively, compared with the untreated I/R group. Histopathological scores for Purkinje cell loss, edema, vascular congestion, and TNF-α expression were also significantly reduced by skimmianine. Quantitative TEM analysis showed that I/R reduced myelin thickness (0.29 ± 0.05 vs. 0.53 ± 0.07 µm), increased G-ratio values (0.75 ± 0.05 vs. 0.63 ± 0.04), and increased vacuolized fibers (24.70 ± 4.20% vs. 3.20 ± 1.10%), whereas skimmianine partially restored myelin thickness (0.42 ± 0.07 µm), reduced the G-ratio (0.68 ± 0.05), and decreased vacuolized fibers (11.20 ± 2.80%; p < 0.05 vs. I/R). Molecular docking demonstrated favorable binding between skimmianine and TNF-α, with a predicted binding energy of −6.953 kcal/mol. Conclusions: These findings indicate that skimmianine exerts neuroprotective effects against secondary cerebellar injury following cerebral I/R through coordinated modulation of oxidative stress, systemic neuroinflammatory responses, astroglial injury-associated pathways, and inflammation-related mechanisms. Full article

Bacterial–fungal interactions are prevalent in microbial communities, and fungi often facilitate bacterial dispersal along networks created by fungal hyphae. Using a microfluidic device, we examined how diverse bacterial species disperse in monoculture versus travel in coculture with Aspergillus flavus. Most of the bacteria traveled further when in coculture, although this was not absolute. Two bacteria showing significant dispersal rates only in coculture were the human pathogens Listeria monocytogenes and Staphylococcus aureus. Mechanistically, L. monocytogenes dispersal required flagella, with dispersal impaired in flagellar mutants but enhanced in ∆mogR strains that upregulate flagellar expression. In contrast, the non-flagellar bacterium S. aureus exhibited a unique, wave-like dispersal pattern along the hyphae, a phenomenon that was abolished in agr quorum-sensing mutants deficient in phenol-soluble modulins (PSMs). In a triculture of L. monocytogenes, S. aureus, and A. flavus, L. monocytogenes limited S. aureus dispersal along the fungal hyphae; however, this inhibition was dependent on an intact L. monocytogenes quorum system. Our findings reveal that bacterial motility on fungal networks arises from diverse, species-specific mechanisms, including flagella, transcriptional regulation, potential quorum-sensing-mediated interactions, as well as other notable dispersal phenomena that warrant further investigation. Full article

Background/Objectives: Uveal melanoma is the most common primary ocular cancer in adults. Patients with metastatic uveal melanoma (mUM) have limited treatment options and poor prognosis. mUM is characterized by high oxidative phosphorylation (OXPHOS), which may be a therapeutic vulnerability for this disease. ONC206 is an imipridone compound that can inhibit OXPHOS indirectly and is currently being evaluated in clinical trials. Thus, we tested the effects of ONC206 on human uveal melanoma cell lines and patient-derived xenografts (PDXs) in vitro and in vivo. Methods: The effects of ONC206 on cell survival, apoptosis, autophagy, oncogenic signaling pathways, and metabolic networks were assessed in vitro using human melanoma cell lines. ONC206 was then tested for safety and anti-tumor activity in vivo using two mUM PDX models. Results: ONC206 treatment produced dose-dependent inhibition of mUM cell growth in vitro, with induction of varying levels of apoptosis and autophagy. ONC206 treatment also downregulated OXPHOS effector proteins and metabolites, thereby impairing mitochondrial OXPHOS. Treatment with ONC206 significantly reduced tumor burden and improved survival in two UM PDX mouse models in vivo. Conclusions: Our findings position ONC206 as a mechanistically distinct agent to target mitochondrial metabolism and to inhibit mUM. As ONC206 is currently being evaluated in multiple clinical studies, our data support further evaluation as a potential new therapeutic strategy for mUM. Full article

Aminochrome, an endogenous neurotoxin, has been implicated in the loss of neuromelanin-containing dopaminergic neurons in the nigrostriatal system in Parkinson’s disease. Although aminochrome-induced oxidative stress and its inhibitory effects on microtubule polymerization are well documented, its impact on protein aggregation remains poorly understood. The aim of this research was to evaluate the effects of aminochrome on protein aggregate accumulation in SH-SY5Y cells differentiated into dopaminergic neurons. While the role of aminochrome in autophagy has been described, its direct effect on autophagosome–lysosome fusion has not been studied. Our findings reveal that aminochrome, like vinblastine, delays autophagosome–lysosome fusion and induces cell death. This inhibitory effect was also observed in the presence of autophagy inducers, which partially attenuated aminochrome-induced cell death. Under these conditions of disruptions in autophagosome–lysosome fusion, a marked accumulation of perinuclear vimentin and ubiquitin aggregates was observed. Aminochrome also increased colocalization between vimentin and ubiquitin. Interestingly, ubiquitin aggregates were also detected within the nucleus. These findings suggest that aminochrome-induced disruption of the microtubule network, particularly its impairment of autophagosome–lysosome fusion and promotion of protein aggregation, may represent a critical mechanism leading to cell death. In addition, inhibition of autophagosome–lysosome fusion may contribute to the accumulation of perinuclear and nuclear protein aggregates, which may be associated with either toxic or non-toxic pathways. Our findings underscore the therapeutic potential of targeting both microtubule stabilization and proteostasis pathways, including autophagy and the ubiquitin–proteasome system (UPS), in Parkinson’s disease, highlighting the need for further research into nuclear proteotoxicity mechanisms. Full article

Background: Several studies have investigated the clinical and immunological aspects of medication-related osteonecrosis of the jaw (MRONJ). However, the underlying immunological mechanisms and signaling pathways involved in its pathophysiology remain incompletely understood. This systematic review and meta-analysis, complemented by bioinformatics analyses, aimed to identify proteomic biomarkers associated with MRONJ. Methods: Six databases (PubMed, Embase, Scopus, Web of Science, Cochrane Library, and VHL) were searched, along with gray literature and manual searches. Observational studies in English comparing proteomic profiles of individuals with and without MRONJ were included. Study selection and data management were conducted using EndNote™ X8 and Rayyan.ai, and risk of bias was assessed using the QUADOMICS tool. Functional enrichment analysis was performed using g:Profiler and Reactome, and interaction networks were constructed using GeneMANIA, STRING, and MetaboAnalyst (Cytoscape program; version 3.10.1). Meta-analysis was performed in RStudio (R-4.5, Rstudio extension 2025.05.1+513) (α = 0.05). Results: Three studies were included in the review, and two in the meta-analysis. The meta-analysis showed higher salivary levels of Apolipoprotein B-100 (APOB), Apolipoprotein A-II (APOA2), and Heparin Cofactor 2 (SERPIND1) in MRONJ patients, while the protein Keratin (KRT16) showed reduced levels without statistical significance. Bioinformatics analyses indicated involvement in lipid metabolism, impaired tissue repair, and inflammatory and immune responses. Conclusions: These findings suggest altered salivary proteomic signatures in MRONJ for APOB, APOA2, SERPIND1, and KRT16 proteins. Full article

Temporal lobe epilepsy (TLE) is a common drug-resistant epilepsy characterized by recurrent seizures, cognitive impairment, aberrant adult hippocampal neurogenesis, inhibitory circuit disruption, and persistent inflammatory remodeling. Current anti-seizure medications primarily offer symptomatic control and do not target the progressive structural and functional deterioration of epileptic hippocampal networks. Here, we investigated whether local nerve growth factor (NGF)-hydrogel delivery during the latent phase after status epilepticus could mitigate hippocampal pathological remodeling and improve long-term outcomes in a kainic acid (KA)-induced mouse model (utilizing C57BL/6J and Nestin-CreERT2 mice). Animals were randomly assigned to three groups: the saline control group, the untreated KA epilepsy group, and the KA + NGF-hydrogel treatment group. NGF-hydrogel was administered into hippocampal Cornu Ammonis 1 (CA1) beginning 3 days post-kainic acid and repeated every 15 days. Histological, immunofluorescence, circuit-tracing, electrophysiology, electroencephalography (EEG), and behavioral assessments were used to evaluate neurogenesis, microenvironment, circuit readouts, seizure burden, and cognition. NGF-hydrogel treatment was associated with preserved dentate gyrus neural stem cell populations, improved newborn granule cell localization and maturation, attenuated neuroinflammation and gliosis, and partial recovery of inhibitory interneuron markers. These changes were accompanied by improved hippocampal circuit readouts, reduced chronic spontaneous seizure burden, and enhanced recognition and spatial memory. Our findings indicate that local NGF-hydrogel delivery following status epilepticus is associated with improved hippocampal remodeling and functional outcomes, and suggest that biomaterial-based neurotrophic support may be a promising strategy for providing targeted neuroprotection and facilitating excitatory/inhibitory (E/I) balance reconstruction in the epileptic hippocampus. Full article

To address the issues of low perception accuracy and poor robustness in traditional motion recognition methods within complex walking environments for visually impaired individuals, this study utilizes multi-modal data, including ECG, PPG, and IMU, for classification. Regarding the low filtering efficiency of multi-modal data, an improved wavelet filtering algorithm based on LSTM is proposed. To further enhance classification accuracy, this paper introduces a motion recognition method for the blindfolded mobility simulation based on an Attention-based Two-Stream Deep Fusion Convolutional Neural Network (ATS-DFCNN). The proposed method constructs a two-stream heterogeneous feature extraction architecture by synchronously collecting tri-axial motion signals and physiological signals from subjects. A 1D-CNN is employed to capture the spatial geometric features of limb movements, while a hybrid CNN-GRU network is utilized to mine the temporal evolution patterns of physiological stress. Furthermore, an attention mechanism is introduced to achieve dynamic weighted fusion at the feature level, which strengthens critical motion features and suppresses environmental noise. Experiments were conducted with 10 subjects simulating the movements of visually impaired individuals, covering typical actions such as walking, standing, climbing stairs, descending stairs, and falling. The results demonstrate that the proposed adaptive filtering algorithm achieves an AUC of 0.942, significantly improving feature distinctiveness compared to traditional algorithms. The ATS-DFCNN model achieved an average recognition accuracy of 92.2% across five activity categories, representing a 4.8% performance increase over single IMU modal classification. Particularly in fall detection, the model effectively reduces false alarms through physiological feedback and accurately infers motion intentions, providing reliable technical support for the safety monitoring of intelligent walking-aid systems. Full article