A5018267773- Dental Radiography and Imaging
- Advanced Image Fusion Techniques
- Image Enhancement Techniques
- Advanced X-ray and CT Imaging
- Dental Implant Techniques and Outcomes
- Topic Modeling
- Infrared Target Detection Methodologies
- Medical Imaging and Analysis
- Natural Language Processing Techniques
- Image and Signal Denoising Methods
- Multimodal Machine Learning Applications
- MXene and MAX Phase Materials
- 3D Shape Modeling and Analysis
- Machine Learning and Data Classification
- Medical Imaging Techniques and Applications
- Endodontics and Root Canal Treatments
- Domain Adaptation and Few-Shot Learning
- Machine Learning and Algorithms
- Advanced Image and Video Retrieval Techniques
- Advanced Measurement and Detection Methods
- Advanced Graph Neural Networks
- Advanced materials and composites
- Nuclear Materials and Properties
- Text and Document Classification Technologies
- Image Processing and 3D Reconstruction
Zhejiang University
2023-2025
Shanghai Advanced Research Institute
2025
Chinese Academy of Sciences
2025
Northwestern Polytechnical University
2024
Stomatology Hospital
2023
University of Illinois Urbana-Champaign
2022-2023
Jiangxi University of Finance and Economics
2021
This paper proposes an effective infrared and visible image fusion method based on a texture conditional generative adversarial network (TC-GAN). The constructed TC-GAN generates combined map for capturing gradient changes in fusion. generator the is designed as codec structure extracting more details, squeeze-and-excitation module applied to this increase weight of significant information map. loss function by combing retain source images. discriminator brings generated closer image. To...
High-fidelity three-dimensional (3D) models of tooth-bone structures are valuable for virtual dental treatment planning; however, they require integrating data from cone-beam computed tomography (CBCT) and intraoral scans (IOS) using methods that either error-prone or time-consuming. Hence, this study presents Deep Dental Multimodal Fusion (DDMF), an automatic multimodal framework reconstructs 3D CBCT IOS. Specifically, the DDMF comprises IOS segmentation modules as well a reconstruction...
The Contrastive Language-Image Pre-Training (CLIP) model, pretrained on large visual text corpora, has demonstrated significant improvements in and linguistic tasks been applied to various downstream tasks. At least two issues, however, hinder the transfer learning with such powerful models field of medical question answering (Med-VQA). That current methods tend full fine-tune these large-scale are suffering from increasingly expensive computational cost as model size grows development....
Environmental contamination by U(VI) from radioactive wastewater has become a challenging concern in the development of nuclear energy. A highly efficient recovery is essential for environmental remediation and can mitigate depletion conventional uranium resources. This study describes synthesis single-layered Ti3C2TX nanosheets chemical exfoliation using ultrasonography. The structure promoted high sorption capacity 3.20 mmol/g (distribution constant Kd > 104 mL/g) excellent selectivity...
Methods of rain removal based on deep learning have rapidly developed, and the image quality after is continuously improving. However, results most methods some common problems, including a loss details, blurring edges, existence artifacts. To remove rain-related information more thoroughly retain edge this paper proposes an end-to-end network progressive residual detail supplement (ERRN-PRDS) approach. The entire structure designed in iterative manner to obtain higher-quality images from...
The fusion of infrared and visible images combines the advantages thermal radiation information in texture images. To preserve salient targets obtain better results, this article proposes a novel image method based on dual-kernel side window filtering detail optimization with S-shaped curve transformation. First, box filter (DSWBF) adaptive kernel size is designed to extract base layers source Then, saliency-based rule proposed for highlight regions image. Next, module constructed an...
Abstract Cone-beam computed tomography (CBCT) is one of the most widely used digital models in dental practices. A critical step virtual treatment planning to accurately delineate all tooth-bone structures from CBCT with high fidelity. Previous studies have established several methods for segmentation using deep learning. However, inherent resolution discrepancy and loss occlusal information largely limited its clinical applicability. Here, we present a Deep Dental Multimodal Analysis (DDMA)...
Mitigating the hallucinations of Large Language Models (LLMs) and enhancing them is a crucial task. Although some existing methods employ model self-enhancement techniques, they fall short effectively addressing unknown factual hallucinations. Using Knowledge Graph (KG) enhancement approaches fails to address generalization across different KG sources open-ended answer questions simultaneously. To tackle these limitations, there framework that combines Pseudo-Graph Generation Atomic...
The Royal Society of Chemistry is the world's leading chemistry community.
Medical Visual Question Answering (MedVQA), which offers language responses to image-based medical inquiries, represents a challenging task and significant advancement in healthcare. It assists experts swiftly interpret images, thereby enabling faster more accurate diagnoses. However, the model interpretability transparency of existing MedVQA solutions are often limited, posing challenges understanding their decision-making processes. To address this issue, we devise semi-automated...