A5100423150- Brain Tumor Detection and Classification
- Advanced Neural Network Applications
- Neuroinflammation and Neurodegeneration Mechanisms
- Medical Image Segmentation Techniques
- Parkinson's Disease Mechanisms and Treatments
- Acute Ischemic Stroke Management
- Image Enhancement Techniques
- Video Surveillance and Tracking Methods
- Neurological disorders and treatments
- Advanced Image and Video Retrieval Techniques
- Satellite Communication Systems
- Advanced Data Compression Techniques
- Telecommunications and Broadcasting Technologies
- Traumatic Brain Injury and Neurovascular Disturbances
- Visual Attention and Saliency Detection
- Cardiovascular Function and Risk Factors
- Advanced Image Processing Techniques
- Biochemical Analysis and Sensing Techniques
- S100 Proteins and Annexins
- Hormonal Regulation and Hypertension
- Adipose Tissue and Metabolism
- Adenosine and Purinergic Signaling
- Ultrasound and Hyperthermia Applications
- Electrophoretic Deposition in Materials Science
- Glaucoma and retinal disorders
Fudan University
2021-2024
Xidian University
2024
Beihang University
2015-2023
First Affiliated Hospital of Dalian Medical University
2020-2022
Third Affiliated Hospital of Southern Medical University
2022
Dalian Medical University
2020-2022
Nanjing Medical University
2022
People's Liberation Army No. 150 Hospital
2021
Chinese People's Liberation Army
2021
Wenzhou Medical University
2020
Abstract Disability in multiple sclerosis (MS) is driven part by the failure of remyelination and progressive neurodegeneration. Microglia, specifically triggering receptor expressed on myeloid cells 2 (TREM2), a factor highly microglia, have been shown to play an important role remyelination. Here, using focal demyelination model brain, we demonstrate that persistent TREM2 knockout mice, lasting more than 6 weeks after lysolecithin injection resulting substantial We also find mice exhibit...
Abstract Purpose To determine the relationships among retinal/choroidal thickness, retinal microvascular network and visual field in high myopia. Methods This cross‐sectional study included a total of 62 subjects, comprising 31 eyes with myopia emmetropia or low Optical coherence tomography was used to quantify thickness ganglion cell complex (GCC), inner nuclear layer outer plexiform (INOPL), (ORL) choroid (ChL). angiography superficial vessel density (SVD) deep (DVD). Retinal light...
Diffusion-weighted magnetic resonance imaging (DWI) is sensitive to acute ischemic stroke and a common diagnostic method for the stroke. However, result relies on visual observation of neurologists which may vary from doctor under different circumstance. And manual segmentation often time-consuming subjective process. The time onset thrombus removal has significant impact prognosis patients with shorter time, better prognosis. For this purpose we present novel framework quickly automatically...
Semantic segmentation in urban scenes is widely used applications of intelligent transportation systems (ITS). In scenes, thermal infrared (TIR) images can be captured weak illumination conditions or the presence obscuration (e.g., light fog, smoke). Therefore, TIR have great potential to endow automated vehicles assist navigation systems. However, imaging blurry and low-contrast due absorption by atmospheric gases heat transfer effect. Hence, semantic has rarely been explored even though it...
<title>Abstract</title> The exponential rise in ransomware attacks has intensified the demand for sophisticated detection methodologies capable of addressing complex evasion tactics. Dynamic Obfuscation Signature Analysis (DOSA) offers an adaptive, multi-layered framework designed to counter ransomware’s polymorphic transformations through a hybrid approach that combines static analysis, dynamic signature mapping, and machine learning-based adaptation. DOSA’s architecture enhances accuracy...
The adaptive immune system has implications in pathology of Parkinson’s disease (PD). Research data demonstrated that the peripheral CD4+ T-cell population decreased pathogenesis PD. effect damaged dopaminergic neurons on T cells PD is still unknown. In this study, we constructed a neuronal and glial co-culture model by using human neuroblastoma SH-SY5Y gliomas U87. After were treated with neurotoxin 1-methyl-4-phenylpyridinium (MPP+) for 24 h, conditioned media was harvested used to...
Intracranial hemorrhage (ICH) classification from computed tomography (CT) scans is important for accurate diagnosis of stroke in emergency centers. However, it challenged by low image contrast and the complex appearance lesion. Therefore, results may vary doctor to under different situations. Existing auxiliary algorithms treat features each location as equally important, they do not impose any constraints on learned features. The used model contain noise or that have no diagnostic...
The Low Earth Orbit (LEO) satellite communication system has become an important trend of next-generation services. As it is known, rain attenuation affects the between and ground, significant to evaluate effect on availability LEO system. In this paper, a link evaluation model built with consideration for On basis, service algorithm proposed using Monte Carlo method. Finally, case used find
Abstract Background Vascular Parkinsonism(VaP) is defined as parkinsonism resulting from cerebral vascular disease(CVD), with presence of variable motor and non-motor signs that are corroborated by clinical, anatomic or imaging findings cerebrovascular disease. Overlapping syndromes mixed pathologies make VaP difficult to distinguish primary neurodegenerative parkinsonism.To understand the clinical pathological features VaP,we report a case autopsy confirmed Parkinsonism was misdiagnosed...
In ultrasound video, the tracking and segmentation of carotid plaques are prerequisite for analysis plaque properties. The problem at hand is quite challenging as some issues such low image resolution, large variation between frames efficiency need to addressed. Our method, Siamese automatic network (SiamATSN), an end-to-end deep learning method. Multiple dual attention region proposal (DARPN) blocks developed integrate multi-level features better target detection. DARPN block composed...
With neural networks growing deeper and feature maps larger, limited communication bandwidth with external memory (or DRAM) power constraints become a bottle-neck in implementing network inference on mobile edge devices. In this paper, we propose an end-to-end differentiable efficient method the activation compressed by data compression method. Specifically, transform-quantization-entropy coding pipeline for symmetric exponential Golomb data-dependent Gaussian entropy model arithmetic...
Based on VOXEL-MAN software and our previous work merging acupoint information with the dataset of male Visible Human, we have refined segmentation method for muscles visualized anatomical structure meridians. In interactive segmentation, obtain additional connectivity texture. We also apply surface fitting to track muscle contours data slices where threshold is hard or impossible apply. With these processes, original method, mainly via threshold, improved deal more complex situations....