Multi-modal Magnetic Resonance Imaging (MRI), which is pivotal for comprehensive brain analysis, frequently suffers from missing modalities in clinical settings. Current synthesis methods, which often produce anatomical hallucinations, fail to preserve 3D structural consistency. To address these issues, we propose Swin-Control-LDM, which is a novel dual-stream latent diffusion framework. We introduce a Global-Local Spatial Adaptive Normalization (GL-SPADE) module, which explicitly disentangles anatomical structure from modality style in the latent space. Furthermore, we construct a 3D Swin-Control backbone, which integrates Swin Transformers for long-range dependency modeling. This backbone leverages a ControlNet with zero-convolution injection, which imposes strict structural constraints. Extensive experiments, which utilize the BraTS 2021, BraTS 2023, and IXI datasets, demonstrate that our method achieves state-of-the-art performance. Our approach generates high-fidelity missing modalities, which maintain anatomically consistent structures.

Epileptic seizure detection from electroencephalogram (EEG) signals is critical for clinical diagnosis and long-term neurological monitoring. However, conventional artificial neural networks (ANNs) are often computationally expensive and energy demanding, which hinders their deployment in large-scale or real-time brain-signal analysis. Spiking neural networks (SNNs) provide a biologically inspired and energy-efficient alternative, yet existing architectures still struggle to balance accuracy and efficiency in EEG-based seizure detection. In this study, we propose an adaptive integrate-and-fire (AIF) spiking neuron model that dynamically adjusts its temporal behavior to capture diverse activation patterns. Based on this neuron, we develop a dual-branch spiking neural network (DBSNet), designed to decode multi-scale and multi-dimensional EEG features for improved seizure detection. We evaluate DBSNet on three public epileptic EEG datasets. Among SNN-based approaches, DBSNet consistently achieves state-of-the-art performance. On a large-scale dataset, it even surpasses the best-performing ANN while consuming only one-seventh of its theoretical energy, highlighting its efficiency advantage. These results demonstrate the potential of adaptive spiking architectures to achieve accurate and sustainable neural computing for EEG-based seizure detection, and they suggest a promising paradigm for broader applications in brain-signal processing.

Cerebral microbleeds (CMBs), detected through susceptibility-weighted imaging (SWI) in magnetic resonance images, are important diagnostic indicators for various cerebrovascular diseases. The detection rate of CMBs generally increases with higher field strengths. This study explored the efficacy of detecting CMBs using the minimum intensity projection (MinIP) technique with 7T ultra-high field MRI. MinIP, a post-processing technique applied to SWI, maintains key characteristics of CMBs on each slice while preserving the continuous structure of blood vessels, offering potential advantages for automated detection. We compared the performance of CMB detection using the YOLO-v7 model with SWI and MinIP as inputs. In tests using the original images (size: 1456x1792 pixels and 1344x1792 pixels), the detection performance of SWI and MinIP was comparable. However, when a sliding window approach (patch size: 640x640 pixels) was employed, MinIP achieved a significant improvement in sensitivity (91.05%) while maintaining comparable precision (93.05%) to SWI, which achieved a 80.32% sensitivity and 95.40% precision. These findings suggest that ultra-highresolution MinIP, particularly when combined with a sliding window technique, can substantially enhance the detection of CMBs in deep learning models.

Determine whether APOE gene polymorphism is associated with hypoperfusion intensity ratio (HIR) in acute ischemic stroke (AIS) patients with large vessel occlusion (LVO). Continuously reviewed hospitalized LVO-AIS patients. According to whether the patients carried APOE allele ε 4, they were divided into 2 groups: ε4 carriers and non-ε4 carriers. CTP assessed HIR and infarct core (IC) volume. Good collaterals were defined as HIR < 0.4 and poor collaterals were defined as HIR ≥ 0.4. The patients were divided into two groups based on their HIR value: the HIR < 0.4 group and the HIR ≥ 0.4 group. IC volume was a rCBF < 40% volume. NIHSS at admission assessed stroke severity. A total of 101 patients with LVO-AIS were enrolled, including 82 patients with HIR < 0.4 and 19 patients with HIR ≥ 0.4. Among the patients with HIR < 0.4, 10 were ε4 carriers (12.20%), while among those with HIR ≥ 0.4, 8 were ε4 carriers (42.11%). The proportion of ε4 carriers was significantly higher in the HIR ≥ 0.4 patients group (P = 0.006). In all enrolled patients, there were 83 non-ε4 carriers and 18 ε4 carriers. The IC volume in ε4 carriers was significantly higher than that in non-ε4 carriers (P = 0.003). The NIHSS score in ε4 carriers was significantly higher than that in non-ε4 carriers (P = 0.004). Binary logistic regression showed that APOE ε4 was an independent risk factor for poor collaterals (OR = 6.00, 95%CI: 1.80, 20.02, P = 0.004). Multiple linear regression showed HIR had a significant positive effect on IC volume (B = 167.70, P < 0.001) and NIHSS score (B = 8.53, P = 0.014). APOE ε4 is an independent risk factor for poor collaterals.

Accurately and automatically segmenting the hippocampus into multiple subfields on magnetic resonance imaging images is crucial for the diagnosis and intervention of various neurological diseases. However, small sizes and complex structures of the hippocampal subfields, along with their unclear boundaries and significant volume imbalances, pose substantial challenges for automatic segmentation. To address these challenges, we propose a novel Multi-level Wavelet Fusion Network (MWFNet) to enhance the segmentation accuracy of hippocampal subfields. MWFNet incorporates multi-level wavelet transforms during the encoding process, effectively supplementing spatial domain information with wavelet domain features to improve the perception of detailed semantic information. Additionally, it recovers information lost due to downsampling, particularly benefiting smaller hippocampal subfields. Additionally, we developed a Multi-scale Attention Residual Block (MARB) that leverages convolutional kernels of different sizes to facilitate multi-scale feature extraction. MARB integrates channel and spatial attention to adaptively extract the most effective image features. Combining MARB, we also introduced a new deep supervision scheme to enhance MWFNet's attention and supervision on effective deep features. Extensive experiments conducted on two public hippocampal subfield datasets show that our approach surpasses other state-of-the-art methods. Specifically, MWFNet achieved an average Dice Similarity Coefficient (DSC) score of 75.27% and an average 95th percentile Hausdorff Distance (HD95) score of 0.96 mm on the PHS dataset; on the UMC dataset, it achieved an average DSC score of 77.66% and an average HD95 score of 1.08 mm. Compared to existing hippocampal subfield segmentation methods, MWFNet represents a significant attempt to incorporate multi-level wavelet transforms into this task and has illustrated superior performance.

Endocrine-disrupting chemicals (EDCs) are exogenous substances discharged into the environment through human activities. 3-Isopropylphenol, a typical alkylphenol-based EDC, has been extensively studied due to its broad application and potential ecological impacts. However, the mechanism of its neurotoxicity remains unclear. In this study, the neurotoxic effects of 3-isopropylphenol were examined using the zebrafish model. We predicted its potential toxic mechanisms and action targets using network toxicology and molecular docking and verified them via RT-qPCR. Results showed that 3-isopropylphenol exposure inhibits the cAMP/PKA signaling pathway in zebrafish larvae, promoting apoptosis, impairing neural development, and suppressing locomotor behavior. These findings enhance our understanding of the toxic effects and mechanisms of 3-isopropylphenol on zebrafish larval neural development and aid in evaluating its potential ecological hazards.

Background:Angiogenesis is vital for tissue repair but insufficient in chronic wounds due to paradoxical growth factor overexpression yet reduced neovascularization. Therapeutics physiologically promoting revascularization remain lacking. This study aims to investigate the molecular mechanisms underlying fibroblast-derived exosome-mediated angiogenesis during wound repair. Methods:To assess the effects of fibroblasts derived exosomes on wound healing and angiogenesis, a full-thickness mouse skin injury model was established, followed by pharmacological inhibition of exosome secretion. The number and state of blood vessels in wounds were assessed by immunofluorescence, immunohistochemistry, hematoxylin-eosin staining, and laser Doppler imaging system. The high-throughput miRNA sequencing was carried out to detect the miRNA profiles of fibroblast-derived exosomes. The roles of candidate miRNAs, their target genes, and relevant pathways were predicted by bioinformatic online software. The knockdown and overexpression of candidate miRNAs, co-culture system, matrigel assay, pharmacological blockade, cell migration, EdU incorporation assay, and cell apoptosis were employed to investigate their contribution to angiogenesis mediated by fibroblast-derived exosomes. The expression of vascular endothelial growth factor A (VEGFA), vascular endothelial growth factor receptor 2 (VEGFR2), hypoxia-inducible factor 1α (HIF-1α), von Hippel-Lindau (VHL), and proline hydroxylases 2 was detected by western blot, co-immunoprecipitation, immunofluorescence, real-time quantitative polymerase chain reaction, flow cytometry, and immunohistochemistry. Furthermore, a full-thickness mouse skin injury model based on type I diabetes mellitus induced by streptozotocin was established for estimating the effect of fibroblast-derived exosomes on chronic wound healing. Results:Pharmacological inhibition of exosome biogenesis markedly reduces neovascularization and delays murine cutaneous wound closure. Topical administration of fibroblast-secreted exosomes rescues these defects. Mechanistically, exosomal microRNA-24-3p suppresses VHL E3 ubiquitin ligase levels in endothelial cells to stabilize hypoxia-inducible factor-1α and heighten vascular endothelial growth factor signaling. MicroRNA-24-3p-deficient exosomes exhibit attenuated pro-angiogenic effects. Strikingly, topical application of exosomes derived from fibroblasts onto chronic wounds in diabetic mice improves neovascularization and healing dynamics. Conclusions:Overall, we demonstrate central roles for exosomal miR-24-3p in stimulating endothelial HIF-VEGF signaling by inhibiting VHL-mediated degradation. The findings establish fibroblast-derived exosomes as promising acellular therapeutic candidates to treat vascular insufficiency underlying recalcitrant wounds.

AIM:Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by significant heterogeneity in clinical symptoms and underlying neurobiology. This study aimed to identify distinct ASD biotypes and uncover their neurobiological underpinnings using a novel graph-based subtyping approach. METHODS:Resting-state fMRI and clinical data from 443 males with ASD (17.22 ± 8.63 years) were analyzed. We proposed a population graph-based dual autoencoder for subtyping (PG-DAS), a deep clustering framework that integrates imaging data and nonimaging data to extract deep features for biotype identification. Statistical analyses were conducted to compare clinical scores and functional connectivity patterns between biotypes. Correlation analyses examined the associations between intra- and internetwork connectivity and clinical symptoms. Predictive modeling using support vector regression assessed the ability of network connectivity to predict clinical scores. RESULTS:Two distinct ASD biotypes were identified. ASD1 exhibited significantly lower clinical scores and reduced network integration, characterized by weaker intra- and internetwork connectivity, particularly in core networks such as the cingulo-opercular network, linked to communication symptom scores. In contrast, ASD2 exhibited greater network segregation, with internetwork connectivity in sensorimotor-related networks correlating with total symptom scores. Predictive modeling further revealed biotype-specific brain-behavior associations, with ASD1 and ASD2 showing positive correlations with social and communication scores, respectively. CONCLUSION:This study underscores the critical role of biotype-specific brain network patterns in understanding ASD heterogeneity. The proposed PG-DAS framework proved effective in ASD subtyping and holds promise for broader applications in exploring other neuroheterogeneous disorders.

The accurate segmentation of brain tumor is significant in clinical practice. Convolutional Neural Network (CNN)-based methods have made great progress in brain tumor segmentation due to powerful local modeling ability. However, brain tumors are frequently pattern-agnostic, i.e. variable in shape, size and location, which can not be effectively matched by traditional CNN-based methods with local and regular receptive fields. To address the above issues, we propose a shape-scale co-awareness network (S2CA-Net) for brain tumor segmentation, which can efficiently learn shape-aware and scale-aware features simultaneously to enhance pattern-agnostic representations. Primarily, three key components are proposed to accomplish the co-awareness of shape and scale. The Local-Global Scale Mixer (LGSM) decouples the extraction of local and global context by adopting the CNN-Former parallel structure, which contributes to obtaining finer hierarchical features. The Multi-level Context Aggregator (MCA) enriches the scale diversity of input patches by modeling global features across multiple receptive fields. The Multi-Scale Attentive Deformable Convolution (MS-ADC) learns the target deformation based on the multiscale inputs, which motivates the network to enforce feature constraints both in terms of scale and shape for optimal feature matching. Overall, LGSM and MCA focus on enhancing the scale-awareness of the network to cope with the size and location variations, while MS-ADC focuses on capturing deformation information for optimal shape matching. Finally, their effective integration prompts the network to perceive variations in shape and scale simultaneously, which can robustly tackle the variations in patterns of brain tumors. The experimental results on BraTS 2019, BraTS 2020, MSD BTS Task and BraTS2023-MEN show that S2CA-Net has superior overall performance in accuracy and efficiency compared to other state-of-the-art methods. Code: https://github.com/jiangyu945/S2CA-Net.

The accurate segmentation of brain tumors is crucial for clinicians to diagnose conditions and plan surgeries. Convolutional neural network (CNN)-based methods have made significant progress in brain tumor segmentation. However, real clinical environments often face issues with missing modalities. To address this, we propose GLR-Net, a network that uses shared representation learning for brain tumor segmentation with missing modalities. GLR-Net learns general latent representations from multi-modal features, which avoids the costs and quality issues of synthesizing missing modalities. The framework includes a multi-encoder-multi-decoder structure with unimodal branches for feature learning and a common multi-modal branch for joint segmentation. A modal correlation learner is also introduced to enhance robustness in scenarios with missing modalities. Experimental results on the BraTS 2019 and BraTS 2020 datasets validate the effectiveness of GLR-Net, which achieves robust brain tumor segmentation even with missing modalities.

Autism spectrum disorder (ASD) is a heterogeneous, early-onset neurodevelopmental condition characterized by persistent impairments in social interaction and communication. This study aims to delineate ASD subtypes based on individual gray matter brain networks and provide new insights from a graph theory perspective. In this study, we extracted and normalized single-subject gray matter networks and calculated each network's topological properties. The heterogeneity through discriminative analysis (HYDRA) method was utilized to subtype all patients based on network properties. Next, we explored the differences among ASD subtypes in terms of network properties and clinical measures. Our investigation identified three distinct ASD subtypes. In the case-control study, these subtypes exhibited significant differences, particularly in the precentral gyrus, lingual gyrus, and middle frontal gyrus. In the case analysis, significant differences in global and nodal properties were observed between any two subtypes. Clinically, subtype 1 showed lower VIQ and PIQ compared to subtype 3, but exhibited higher scores in ADOS-Communication and ADOS-Total compared to subtype 2. The results highlight the distinct brain network properties and behaviors among different subtypes of male patients with ASD, providing valuable insights into the neural mechanisms underlying ASD heterogeneity.

Background Severe burns may alter the stability of the intestinal flora and affect the patient's recovery process. Understanding the characteristics of the gut microbiota in the acute phase of burns and their association with phenotype can help to accurately assess the progression of the disease and identify potential microbiota markers.Methods We established mouse models of partial thickness deep III degree burns and collected faecal samples for 16 S rRNA amplification and high throughput sequencing at two time points in the acute phase for independent bioinformatic analysis.Results We analysed the sequencing results using alpha diversity, beta diversity and machine learning methods. At both time points, 4 and 6 h after burning, the Firmicutes phylum content decreased and the content of the Bacteroidetes phylum content increased, showing a significant decrease in the Firmicutes/Bacteroidetes ratio compared to the control group. Nine bacterial genera changed significantly during the acute phase and occupied the top six positions in the Random Forest significance ranking. Clustering results also clearly showed that there was a clear boundary between the communities of burned and control mice. Functional analyses showed that during the acute phase of burn, gut bacteria increased lipoic acid metabolism, seleno-compound metabolism, TCA cycling, and carbon fixation, while decreasing galactose metabolism and triglyceride metabolism. Based on the abundance characteristics of the six significantly different bacterial genera, both the XGboost and Random Forest models were able to discriminate between the burn and control groups with 100% accuracy, while both the Random Forest and Support Vector Machine models were able to classify samples from the 4-hour and 6-hour burn groups with 86.7% accuracy.Conclusions Our study shows an increase in gut microbiota diversity in the acute phase of deep burn injury, rather than a decrease as is commonly believed. Severe burns result in a severe imbalance of the gut flora, with a decrease in probiotics and an increase in microorganisms that trigger inflammation and cognitive deficits, and multiple pathways of metabolism and substance synthesis are affected. Simple machine learning model testing suggests several bacterial genera as potential biomarkers of severe burn phenotypes.

Motor imagery electroencephalogram (EEG)-based brain-computer interfaces (BCIs) offer significant advantages for individuals with restricted limb mobility. However, challenges such as low signal-to-noise ratio and limited spatial resolution impede accurate feature extraction from EEG signals, thereby affecting the classification accuracy of different actions. To address these challenges, this study proposes an end-to-end dual-branch network (EEG-DBNet) that decodes the temporal and spectral sequences of EEG signals in parallel through two distinct network branches. Each branch comprises a local convolutional block and a global convolutional block. The local convolutional block transforms the source signal from the temporal-spatial domain to the temporal-spectral domain. By varying the number of filters and convolution kernel sizes, the local convolutional blocks in different branches adjust the length of their respective dimension sequences. Different types of pooling layers are then employed to emphasize the features of various dimension sequences, setting the stage for subsequent global feature extraction. The global convolution block splits and reconstructs the feature of the signal sequence processed by the local convolution block in the same branch and further extracts features through the dilated causal convolutional neural networks. Finally, the outputs from the two branches are concatenated, and signal classification is completed via a fully connected layer. Our proposed method achieves classification accuracies of 85.84 Competition 4-2a and BCI Competition 4-2b datasets, respectively, surpassing existing state-of-the-art models. The source code is available at https://github.com/xicheng105/EEG-DBNet.

BACKGROUND:Bacterial infection can delay wound healing and is therefore a major threat to public health. Although various strategies have been developed to treat bacterial infections, antibiotics remain the best option to combat infections. The inclusion of growth factors in the treatment approach can also accelerate wound healing. The co-delivery of antibiotics and growth factors for the combined treatment of wounds needs further investigation.OBJECTIVE:Here we aimed to develop antibiotic and growth factor co-loaded nanoparticles (NPs) to treat Staphylococcus aureus-infected wounds.METHODS:By using our previously prepared reactive oxygen species-responsive material (Oxi-αCD), roxithromycin (ROX)-loaded NPs (ROX/Oxi-αCD NPs) and recombinant human epidermal growth factor (rhEGF)/ROX co-loaded NPs (rhEGF/ROX/Oxi-αCD NPs) were successfully fabricated. The in vivo efficacy of this prepared nanomedicine was evaluated in mice with S. aureus-infected wounds.RESULTS:ROX/Oxi-αCD NPs and rhEGF/ROX/Oxi-αCD NPs had a spherical structure and their particle sizes were 164 ± 5 nm and 190 ± 8 nm, respectively. The in vitro antibacterial experiments showed that ROX/Oxi-αCD NPs had a lower minimum inhibitory concentration than ROX. The in vivo animal experiments demonstrated that rhEGF/ROX/Oxi-αCD NPs could significantly accelerate the healing of S. aureus-infected wounds as compared to the free ROX drug and ROX/Oxi-αCD NPs (P < 0.05).CONCLUSION:ROX and rhEGF co-loaded NPs can effectively eliminate bacteria in wounds and accelerate wound healing. Our present work could provide a new strategy to combat bacteria-infected wounds.

Cell migration is crucial for numerous physiological and pathological processes. A cell adapts its morphology, including the overall and nuclear morphology, in response to various cues in complex microenvironments, such as topotaxis and chemotaxis during migration. Thus, the dynamics of cellular morphology can encode migration strategies, from which diverse migration mechanisms can be inferred. However, deciphering the mechanisms behind cell migration encoded in morphology dynamics remains a challenging problem. Here, we present a powerful universal metric, the Cell Morphological Entropy (CME), developed by combining parametric morphological analysis with Shannon entropy. The utility of CME, which accurately quantifies the complex cellular morphology at multiple length scales through the deviation from a perfectly circular shape, is illustrated using a variety of normal and tumor cell lines in different in vitro microenvironments. Our results show how geometric constraints affect the MDA-MB-231 cell nucleus, the emerging interactions of MCF-10A cells migrating on collagen gel, and the critical transition from proliferation to invasion in tumor spheroids. The analysis demonstrates that the CME-based approach provides an effective and physically interpretable tool to measure morphology in real-time across multiple length scales. It provides deeper insight into cell migration and contributes to the understanding of different behavioral modes and collective cell motility in more complex microenvironments.

Hereditary spastic paraplegia (HSP) is a category of neurodegenerative illnesses with significant clinical and genetic heterogeneity. Homozygous truncated variants of the ERLIN2 gene lead to HSP18 (MIM #611225). However, it is still unclear whether there is an autosomal dominant pathogenic pattern. The specific molecular mechanism needs to be investigated. We generated patient-derived iPSC models to study the mechanism of ERLIN2 heterogeneous variants leading to HSP. We identified a heterozygous missense variant p.Val71Ala of ERLIN2 in an HSP family. Based on IP-mass spectrometry, we found that the ERLIN2 heterozygous missense variant protein recruited the ubiquitin E3 ligase RNF213 to degrade IP3R1. The degradation of IP3R1 leads to the reduction of intracellular free calcium, which triggered endoplasmic reticulum (ER) stress-mediated apoptosis. Calcium homeostasis imbalance inhibited the MAPK signaling pathway that contributed to decreased cell proliferation. In summary, these results suggest that the autosomal dominant inheritance of heterozygous missense variants in ERLIN2 is a novel pathogenic mode of HSP. Furthermore, the disruption of intracellular calcium homeostasis is the pathological mechanism.

Precise delineation of hippocampus subfields is crucial for the identification and management of various neurological and psychiatric disorders. However, segmenting these subfields automatically in routine 3T MRI is challenging due to their complex morphology and small size, as well as the limited signal contrast and resolution of the 3T images. This research proposes Syn_SegNet, an end-to-end, multitask joint deep neural network that leverages ultrahigh-field 7T MRI synthesis to improve hippocampal subfield segmentation in 3T MRI. Our approach involves two key components. First, we employ a modified Pix2PixGAN as the synthesis model, incorporating self-attention modules, image and feature matching loss, and ROI loss to generate high-quality 7T-like MRI around the hippocampal region. Second, we utilize a variant of 3D-U-Net with multiscale deep supervision as the segmentation subnetwork, incorporating an anatomic weighted cross-entropy loss that capitalizes on prior anatomical knowledge. We evaluate our method on hippocampal subfield segmentation in paired 3T MRI and 7T MRI with seven different anatomical structures. The experimental findings demonstrate that Syn_SegNet's segmentation performance benefits from integrating synthetic 7T data in an online manner and is superior to competing methods. Furthermore, we assess the generalizability of the proposed approach using a publicly accessible 3T MRI dataset. The developed method would be an efficient tool for segmenting hippocampal subfields in routine clinical 3T MRI.

Objectives: This study investigated the association between perivascular fat density (PFD) via preoperative computed tomographic angiography (CTA) and early in-stent restenosis (ISR) after carotid artery stenting (CAS). Methods: We retrospectively evaluated 248 consecutive patients who had undergone initial CAS and received a preoperative cervical CTA examination between January 2019 and October 2020. The patients were categorized into two according to whether they sustained ISR during the 2 years postoperative follow-up period. Correlations between PFD and ISR were assessed, and multivariate regression for evaluating predictors of ISR was conducted. Receiver operating characteristic (ROC) curves were used to determine the cutoff value for the PFD. Results: A total of 181 eligible patients (mean age 61.25 & PLUSMN; 10.35 years, 57 male) were enrolled. The ISR group had a higher proportion of closed-cell stents (48.8% versus 27.5%; p = 0.009) and a greater degree of residual stenosis (28[20,33] % versus 20[14.75,30] %; p < 0.001) than the non-ISR group. The ISR group had a higher mean HU value of PFD than the non-ISR group on the operated side (-42.26 & PLUSMN; 6.81 versus-59.66 & PLUSMN; 10.75; p < 0.001). The degree of residual stenosis (OR 1.146, 95%CI 1.071-1.226, p < 0.001) and PFD on the operated side (OR1.353, 95%CI 1.215-1.506, p < 0.001) were significantly associated with the ISR. Conclusions: The occurrence of the early ISR after CAS is associated with a higher PFD on the operated side. The results indicate that PFD is a promising marker to predict the ISR after CAS.