A5028085697- Diabetes Management and Research
- Diabetes and associated disorders
- Pancreatic function and diabetes
- Spinal Cord Injury Research
- Neuroinflammation and Neurodegeneration Mechanisms
- Nerve injury and regeneration
- Galectins and Cancer Biology
- Face recognition and analysis
- EEG and Brain-Computer Interfaces
- Generative Adversarial Networks and Image Synthesis
- Diabetes Treatment and Management
- Neurogenesis and neuroplasticity mechanisms
- Diet and metabolism studies
- 3D Shape Modeling and Analysis
- Spinal Dysraphism and Malformations
- Robotics and Sensor-Based Localization
- Anomaly Detection Techniques and Applications
- Advanced Neural Network Applications
- Facial Nerve Paralysis Treatment and Research
- Industrial Vision Systems and Defect Detection
- Data-Driven Disease Surveillance
- Gut microbiota and health
- Diet, Metabolism, and Disease
- Cervical and Thoracic Myelopathy
- Artificial Intelligence in Healthcare and Education
Second Hospital of Anhui Medical University
2023-2025
Anhui Medical University
2023-2025
Suzhou Municipal Hospital
2024
Second Xiangya Hospital of Central South University
2022-2024
Central South University
2022-2024
Beihang University
2024
Xi’an Jiaotong-Liverpool University
2023-2024
University of Liverpool
2024
Beijing Jiaotong University
2021
Soochow University
2013-2019
Sleep stage classification is critical for diagnosing sleep quality. While deep neural networks are becoming popular automatic with supervised learning, large-scale labeled datasets still hard to acquire. Recently, self-supervised learning (SSL) has become one of the most prominent approaches alleviate burden labeling works. However, existing SSL methods mainly designed non-temporally correlated data and only representations from an instance level. Hence, objective this paper learn robust...
Instance segmentation on point clouds is crucially important for 3D scene understanding. Most SOTAs adopt distance clustering, which typically effective but does not perform well in segmenting adjacent objects with the same semantic label (especially when they share neighboring points). Due to uneven distribution of offset points, these existing methods can hardly cluster all instance points. To this end, we design a novel divide-and-conquer strategy named PBNet that binarizes each and...
Recent studies have confirmed the critical role of neonatal microglia in wound healing and axonal regeneration following spinal cord injury (SCI). However, limited migration to center adult lesion may significantly impede their potential benefits. We established a model microglial centripetal prolonged retention C57BL/6J transgenic mice by injecting exogenous C-X3-C motif chemokine ligand 1 (CX3CL1) macrophage colony-stimulating factor (M-CSF) directly into site post-SCI. Wound...
The purpose of this paper is to learn efficient representations from raw electroencephalogram (EEG) signals for sleep stage classification via self-supervised learning (SSL). Although supervised methods have gained favorable performance, they heavily rely on manually labeled datasets. Recently, SSL arrives comparable performance with fully despite limited data by extracting high-level semantic representations. To alleviate the severe reliance labels, we propose SleepDPC, a novel algorithm...
Diabetes is a major health problem and metabolically unhealthy an important risk factor.To conduct the first nationally representative study on epidemiological data of normal weight (MUNW) focused only nondiabetic subjects determine predictive effect diabetes in China.A longitudinal was conducted using from Rich Healthcare Group China. Metabolic status determined by revised National Cholesterol Education Program Adult Treatment Panel III criteria, individuals with 2 or more criteria were...
Abstract After spinal cord injury (SCI), the accumulation of myelin debris at lesion exacerbates cell death and hinders axonal regeneration. Transplanted bone marrow mesenchymal stem cells (BMSCs) have been proven to be beneficial for SCI repair, but they are susceptible apoptosis. It remains unclear whether this apoptotic process is influenced by debris. Here, we constructed rat BMSCs overexpressing human B-cell lymphoma 2 (hBcl2) alone (hBcl2 group), hBcl2 with an endoplasmic...
With the boom of Large Language Models (LLMs), research solving Math Word Problem (MWP) has recently made great progress. However, there are few studies to examine robustness LLMs in math ability. Instead attacking prompts use LLMs, we propose a MathAttack model attack MWP samples which closer essence problems. Compared traditional text adversarial attack, it is essential preserve mathematical logic original MWPs during attacking. To this end, logical entity recognition identify entries then...
Glycemic variability (GV) in some patients with type 1 diabetes (T1D) remains heterogeneous despite comparable clinical indicators, and whether other factors are involved is yet unknown. Metabolites the serum indicate a broad effect of GV on cellular metabolism therefore more likely to metabolic dysregulation associated T1D. To compare metabolomic profiles between high (GV-H, coefficient variation (CV) glucose ≥ 36%) low (GV-L, CV < groups identify potential biomarkers, metabolomics...
In the context of spinal cord injury (SCI), infiltrating macrophages assume prominence as primary inflammatory cells within lesion core, where fibrotic scar is predominantly orchestrated by platelet-derived growth factor receptor beta (PDGFRβ
Following spinal cord injury (SCI), fibrotic scar inhibits axon regeneration and impairs neurological function recovery. It has been reported that T cell-derived interferon (IFN)-γ plays a pivotal role in promoting scarring neurodegenerative disease. However, the of IFN-γ formation after SCI not declared. In this study, crush mouse was established. Western blot immunofluorescence showed surrounded by fibroblasts at 3, 7, 14, 28 days post-injury. Moreover, is mainly secreted cells SCI. Further,
Objective Regarding the effects and practical application of insulin pumps on patients with type 1 diabetes mellitus (T1DM), real-world evidence is limited especially concerning incidence hypoglycemia. This study aimed to compare efficacy continuous subcutaneous infusion (CSII) therapy multiple daily injection (MDI) glycemic metrics evaluated by retrospective glucose monitoring (CGM) in Chinese T1DM. Methods In total, 362 T1DM from outpatient department Second Xiangya Hospital, Central South...
Abstract Aims To investigate glycaemic variability (GV) patterns in patients with type 1 diabetes (T1D), 2 (T2D), and latent autoimmune adults (LADA). Materials Methods A total of 842 subjects (510 T1D, 105 LADA, 227 T2D) were enrolled underwent week continuous glucose monitoring (CGM). Clinical characteristics CGM parameters compared among T2D. LADA divided into two subgroups based on glutamic acid decarboxylase autoantibody titres (≥180 U/mL [LADA‐1], <180 [LADA‐2]) compared. The...
The comorbidity of metabolic syndrome (MetS) and type 1 diabetes mellitus (T1DM) is an obstacle to glucose control in patients with T1DM. We compared glycemic profiles using continuous monitoring (CGM) systems T1DM or without MetS.This was a multicenter cross-sectional study (N = 207) MetS. CGM data were collected from enrollment until discharge during 1-week session. analyzed baseline HbA1c, average glucose, estimated time range (TIR), above (TAR), below (TBR), coefficient variation (CV),...
We consider the problem of Multi-view 3D Face Reconstruction (MVR) with weakly supervised learning that leverages a limited number 2D face images (e.g. 3) to generate high-quality model very light annotation. Despite their encouraging performance, present MVR methods simply concatenate multi-view image features and pay less attention critical areas eye, brow, nose, mouth). To this end, we propose novel called Deep Fusion (DF-MVR) design encoding single decoding framework skip connections,...
While weakly supervised multi-view face reconstruction (MVR) is garnering increased attention, one critical issue still remains open: how to effectively fuse multiple image information reconstruct high-precision 3D models. In this regard, we propose a novel model called Deep Fusion MVR (DF-MVR) facial shapes from images. Specifically, introduce MulEn-Unet, encoding single decoding framework with skip connections and attention. This design allows for the extraction, integration, compensation...
Abstract Background In the context of spinal cord injury (SCI), infiltrating macrophages assume prominence as primary inflammatory cells within lesion core, where fibrotic scar is predominantly orchestrated by platelet-derived growth factor receptor beta (PDGFRβ + ) fibroblasts. Galectin-3, a carbohydrate-binding protein lectin family, notably expressed hematogenous and mediates cell-cell interactions. Although Galectin-3 has been shown to contribute endocytic internalization PDGFRβ in...
Background: Infiltrating Galectin-3+ macrophages assume prominence as the primary inflammatory cells at lesion core, where fibrotic scar is predominantly orchestrated by platelet-derived growth factor receptor beta (PDGFRβ+) fibroblasts after spinal cord injury (SCI). Although Galectin-3, a carbohydrate-binding protein, mediates cell-cell interactions, its specific role in activating PDGFRβ and driving formation post-SCI remains undetermined.Methods: We employed crush mid-thoracic (T10) SCI...
ABSTRACT Background Effective clearance of lipid‐rich debris by macrophages is critical for neural repair and regeneration after spinal cord injury (SCI). Interleukin‐3 (IL‐3) has been implicated in programming microglia to cluster clear pathological aggregates neurodegenerative disease. Yet, the influence IL‐3 on lipid post‐SCI not well characterized. Methods We established a mouse model compression investigate role IL‐3. Blockage was achieved through intrathecal delivery an...
To the Editor: Type 1 diabetes mellitus (T1DM) is a progressive disease caused by severe islet β-cell function impairment, resulting in high glucose variability (GV) patients.[1] While intensive insulin treatment enabled T1DM patients to achieve near-normal glucose, reflected reduced hemoglobin A1c (HbA1c) levels, stricter HbA1c targets were associated with an increased risk of hypoglycemia, which would compromise patients' quality life and survival.[2] Recent studies suggest that GV...