A5115076721- Topic Modeling
- Adversarial Robustness in Machine Learning
- Natural Language Processing Techniques
- Data Mining Algorithms and Applications
- Spinal Fractures and Fixation Techniques
- Machine Learning and Data Classification
- Cervical and Thoracic Myelopathy
- Domain Adaptation and Few-Shot Learning
- Big Data and Business Intelligence
- Spine and Intervertebral Disc Pathology
- Orthodontics and Dentofacial Orthopedics
- Big Data Technologies and Applications
- AI and Big Data Applications
- Engineering and Test Systems
- Facial Rejuvenation and Surgery Techniques
- Image and Object Detection Techniques
- Advanced Malware Detection Techniques
- Data Stream Mining Techniques
- Periodontal Regeneration and Treatments
- Magnesium Oxide Properties and Applications
- Topology Optimization in Engineering
- Adaptive Control of Nonlinear Systems
- Advanced Vision and Imaging
- Advanced Clustering Algorithms Research
- Advanced oxidation water treatment
Ningbo University
2023-2025
State Grid Corporation of China (China)
2023-2024
Merck (Singapore)
2024
Hong Kong Polytechnic University
2024
First Hospital of Jiaxing
2024
Jiaxing University
2024
Taiyuan University of Science and Technology
2024
Weatherford College
2024
Beijing Institute of Technology
2024
Guangxi University
2024
Background: The optimal PECD surgical approach for cervical intervertebral disc herniation (CIVDH) remains controversial. conventional posterior K-hole leads to damage of facet joint.Objectives: This article is first describe a novel lamina–hole percutaneous endoscopic discectomy (PECD) CIVDH. objective this study evaluate the feasibility and short-term clinical effect approach.Methods: Single-center retrospective observational all patients managed with (PPECD) using symptomatic single-level...
Background: Dementia is a major public health challenge for aging societies worldwide. Neuroinflammation thought to be key factor in dementia development. The aim of this study was comprehensively assess translocator protein (TSPO) expression by positron emission tomography (PET) imaging reveal the characteristics neuroinflammation dementia. Methods: We used meta-analysis retrieve literature on TSPO using PET technology, including but not limited quality design, sample size, and type ligand...
Abstract Background To develop a fully automated CNN detection system based on magnetic resonance imaging (MRI) for ACL injury, and to explore the feasibility of injury MRI images. Methods Including 313 patients aged 16 – 65 years old, raw data are 368 pieces with injured 100 intact ACL. By adding flipping, rotation, scaling other methods expand data, final set is 630 including 355 275 Using proposed model two attention mechanism modules, sets trained tested fivefold cross-validation....
Clustering is an unsupervised machine learning technique whose goal to cluster unlabeled data. But traditional clustering methods only output a set of results and do not provide any explanations the results. Although in literature number based on decision tree have been proposed explain results, most them some disadvantages, such as too many branches deep leaves, which lead complex make it difficult for users understand. In this paper, hypercube overlay model multi-objective optimization...
Computer vision enhanced automatic routing inspection and monitoring for railway pantograph headlines is a promising technical orientation to reduce manual operations. The based approach, which comply exactly with human cognition, draws many researcher's attention because of its informative nature. However, due the dimension collapse in photogenic process that essentially an ill-posed problem, automatically detect locate abnormal visual pattern structure videos still challenging task. We...
In the current study, we restricted our focus on a modified approach to address poor curved attachment of expanded polytetrafluoroethylenein chin augmentation.The implant is shaped generally in accordance with chin, followed by 5-to-8 longitudinally parallel "V"-grooves carved median posterior site where directly attaches mandible. Thus, it enhances bend ductility prosthesis and renders better from surface mandibular embedded region.This procedure was performed 15 patients. After follow-up...
The large-scale multi-objective optimization algorithm (LSMOA), based on the grouping of decision variables, is an advanced method for handling high-dimensional variables. However, in practical problems, interaction among variables intricate, leading to large group sizes and suboptimal effects; hence a weighted overlapping (MOEAWOD) proposed this paper. Initially, are perturbed categorized into convergence diversity variables; subsequently, subdivided groups interactions different If size...
Object detectors frequently encounter significant performance degradation when confronted with domain gaps between collected data (source domain) and from real-world applications (target domain). To address this task, numerous unsupervised adaptive have been proposed, leveraging carefully designed feature alignment techniques. However, these techniques primarily align instance-level features in a class-agnostic manner, overlooking the differences extracted different categories, which results...
Uplift modeling has been widely employed in online marketing by predicting the response difference between treatment and control groups, so as to identify sensitive individuals toward interventions like coupons or discounts. Compared with traditional \textit{conversion uplift modeling}, \textit{revenue modeling} exhibits higher potential due its direct connection corporate income. However, previous works can hardly handle continuous long-tail distribution revenue modeling. Moreover, they...
This paper constructs a multi-source information resource management framework in the cloud computing environment. The system designs from aspects of database structure design, data dictionary optimization, reasonable planning types, index storage procedure writing and calling, etc. In simulation experiment, service strategies under different user scales are compared analyzed. results show that when size trader increases, method can make both parties involved transaction obtain higher...
In autonomous driving, 3D LiDAR plays a crucial role in understanding the vehicle's surroundings. However, newly emerged, unannotated objects presents few-shot learning problem for semantic segmentation. This paper addresses limitations of current segmentation by exploiting temporal continuity data. Employing tracking model to generate pseudo-ground-truths from sequence frames, our method significantly augments dataset, enhancing model's ability learn on novel classes. this approach...
Tabular-format data is widely adopted in various real-world applications. Various machine learning models have achieved remarkable success both industrial applications and data-science competitions. Despite these successes, most current methods for tabular lack accurate confidence estimation, which needed by some high-risk sensitive such as credit modeling financial fraud detection. In this paper, we study the estimation of applied to data. The key finding our paper that a dataset typically...
Current dialogue systems face diverse user requests and rapid change domains, making quickly adapt to scenarios with previous unseen slot types become a major challenge. Recently, researchers have introduced novel detection (NSD) discover potential new types. However, system NSD does not bring practical improvements due the still cannot handle slots in subsequent interactions. In this paper, we define incremental (INSD), which separates deal as two phrases: 1) model discovers unknown slots,...
This paper introduces the development of existing SNMP, and disadvantage current SNMP system. After that system is designing developing, can simultaneously support snmpv1. snmpv2c, snmpv3, be used in a mixed environment ipv4/ipv6. light other similar uses LEAPS algorithm to judge state network, thus change frequency data collection, at same time it based on modular design, upgraded easily, improves systempsilas reusability.
Hidden failures in relay may cause invalidation of protection equipment so that system faults cannot be detected.These will lead to cascading outages severe cases.At present, effective methods lack monitoring hidden failures.A hybrid state estimation which uses data collected from both SCADA and WAMS is proposed.The value obtained regarded as reference value.The measurement uploaded by information compared with value.If the difference exceeds default threshold value, failure can proved exist...