A5100779639- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Multimodal Machine Learning Applications
- Anomaly Detection Techniques and Applications
- Advanced Vision and Imaging
- Text and Document Classification Technologies
- Topic Modeling
- Face and Expression Recognition
- Adenosine and Purinergic Signaling
- Machine Learning and Data Classification
- Robotics and Sensor-Based Localization
- Synthesis and Biological Evaluation
- Natural Language Processing Techniques
- Macrophage Migration Inhibitory Factor
- Computer Graphics and Visualization Techniques
- Non-Invasive Vital Sign Monitoring
- Lung Cancer Diagnosis and Treatment
- Radiomics and Machine Learning in Medical Imaging
- Human Resource Development and Performance Evaluation
- COVID-19 diagnosis using AI
- Cloud Data Security Solutions
- Patient-Provider Communication in Healthcare
- Virus-based gene therapy research
- Healthcare cost, quality, practices
Wenzhou Medical University
2024
The First People's Hospital of Xiaoshan District, Hangzhou
2024
Sun Yat-sen University
2023
Nanyang Technological University
2022
University of Alberta
2022
Chongqing University
2020-2022
Zhengzhou Normal University
2021
Dalian University of Technology
2011
Jingchu University of Technology
2009
Ricoh (Japan)
2007
Compared with single-label image classification, multi-label classification is more practical and challenging. Some recent studies attempted to leverage the semantic information of categories for improving performance. However, these semantic-based methods only take as type complements visual representation without further exploitation. In this paper, we present an innovative path towards solution which considers it a dictionary learning task. A novel end-to-end model named Deep Semantic...
Homologous recombination deficiency (HRD) has emerged as a critical prognostic and predictive biomarker in oncology. However, current testing methods, especially those reliant on targeted panels, are plagued by inconsistent results from the same samples. This highlights urgent need for standardized benchmarks to evaluate HRD assay performance. In phases IIa IIb of Chinese Harmonization Project, we developed ten pairs well-characterized DNA reference materials derived lung, breast, melanoma...
The CAR T-cell therapy is a promising approach to treating hematologic malignancies. However, the application in solid tumors still has many tough challenges, including heterogenicity antigen expressions and immunosuppressive tumor microenvironment (TME). As new cancer treatment modality, oncolytic virotherapy can be engineered circumvent these obstacles for T cell tumors.In this study, an oHSV T7011 drive ectopic expression of dual-antigens, extracellular domains CD19 BCMA, on surface...
Facial image-based kinship verification is a rapidly growing field in computer vision and biometrics. The key to determining whether pair of facial images has kin relation train model that can enlarge the margin between faces have no while reducing distance relation. Most existing approaches primarily exploit duplet (i.e., two input samples without cross pair) or triplet single negative for each positive with low-order information, omitting discriminative features from multiple pairs. These...
In this work, we propose a novel generative approach named Robust Bidirectional Generative Network (RBGN) based on Conditional Adversarial (CGAN) for Generalized Zero-shot Learning (GZSL). RBGN employs the adversarial attack to train more rigorous discriminator, thus enhancing generalizability and robustness of feature generator under minimax strategy. Moreover, decodes generated visual features back their semantic representations further improve representational ability alleviate hubness...
The current LiDAR SLAM (Simultaneous Localization and Mapping) system suffers greatly from low accuracy limited robustness when faced with complicated circumstances. From our experiments, we find that systems have performance the noise level in obtained point clouds is large. Therefore, this work, propose a general framework to tackle problem of denoising loop closure for complex environments many noises outliers caused by reflective materials. Current approaches are mainly designed...
Vision-Language Pre-training has demonstrated its remarkable zero-shot recognition ability and potential to learn generalizable visual representations from language supervision. Taking a step ahead, language-supervised semantic segmentation enables spatial localization of textual inputs by learning pixel grouping solely image-text pairs. Nevertheless, the state-of-the-art suffers clear gaps between modality: plenty concepts appeared in images are missing their paired captions. Such...
This article describes the implementation of Word Sense Disambiguation system that participated in SemEval-2007 multilingual Chinese-English lexical sample task. We adopted a supervised learning approach with Maximum Entropy classifier. The features used were neighboring words and their part-of-speech, as well single context, other syntactic based on shallow parsing. In addition, we word category information Chinese thesaurus for verb disambiguation. For task in, obtained precision 0.716...
The recent Segment Anything Model (SAM) has demonstrated remarkable zero-shot capability and flexible geometric prompting in general image segmentation. However, SAM often struggles when handling various unconventional images, such as aerial, medical, non-RGB images. This paper presents CAT-SAM, a ConditionAl Tuning network that adapts toward target tasks with just few-shot samples. CAT-SAM freezes the entire its mask decoder encoder simultaneously small number of learnable parameters. core...
Three-dimensional perception from multi-view cameras is a crucial component in autonomous driving systems, which involves multiple tasks like 3D object detection and bird's-eye-view (BEV) semantic segmentation. To improve precision, large image encoders, high-resolution images, long-term temporal inputs have been adopted recent models, bringing remarkable performance gains. However, these techniques are often incompatible training inference scenarios due to computational resource...
In November 2023, our hospital confirmed a case of Mycolicibacterium mucogenicum infection in the right back. The patient sought medical attention at due to "a back lump for 2 months, ruptured 1 month." Two months ago, patient's collided with fire hose and developed local walnut sized lump. was soft, painless when pressed, there no obvious redness or swelling around it. went private incision drainage treatment on their own. One month later, condition did not improve yellow pus exuding from...
The connection between emotional states and physical health has attracted widespread attention. stress assessment can help healthcare professionals figure out the patient's engagement toward diagnostic plan optimize rehabilitation program as feedback. It is of great significance to study changes physiological features in process change find subset one or several that best represent psychological state a statistical sense. Previous studies had used differences discrete select feature subsets....
Cavitary lung nodules are the main hindrance to clinical application of lung-nodule-segmentation methods. In this study, avoid disturbance high granularity in computed tomography images, we design a novel Dual Encoding Fusion Network (DEF-Net) for atypical nodule segmentation. DEF-Net is composed three parts: multi-level representations, dual encoding, and fusion decoding. The global local characteristics can be better captured by fusing two encodings original inputs. Furthermore, dilated...
In this paper, we describe the participation of team ”CLP” in SemEval-2019 Task 3 “Con- textual Emotion Detection Text” that aims to classify emotion user utterance tex- tual conversation. The submitted system is a deep learning architecture based on Hier- archical Attention Networks (HAN) and Em- bedding from Language Model (ELMo). core contains two represen- tation layers. first one combines out- puts ELMo, hand-craft features Bidi- rectional Long Short-Term Memory with At- tention...
<div>Abstract<p>Although tyrosine kinase inhibitor therapy and immunotherapy have significantly improved lung cancer management, many patients do not benefit or become resistant to treatment, highlighting the need for novel treatments. We found elevated CD73 expression be prevalent in non–small cell (NSCLC) including those harboring RAS- RTK (EGFR, EML4-ALK) oncogenes. is enriched closely with transcriptome signature of epithelial–mesenchymal transition immune-tolerant tumor...
<div>Abstract<p>Although tyrosine kinase inhibitor therapy and immunotherapy have significantly improved lung cancer management, many patients do not benefit or become resistant to treatment, highlighting the need for novel treatments. We found elevated CD73 expression be prevalent in non–small cell (NSCLC) including those harboring RAS- RTK (EGFR, EML4-ALK) oncogenes. is enriched closely with transcriptome signature of epithelial–mesenchymal transition immune-tolerant tumor...