A5103814778- Medical Imaging Techniques and Applications
- Domain Adaptation and Few-Shot Learning
- Radiomics and Machine Learning in Medical Imaging
- Advanced X-ray and CT Imaging
- Nuclear Physics and Applications
- Advanced Radiotherapy Techniques
- Radiation Detection and Scintillator Technologies
- COVID-19 diagnosis using AI
- Face and Expression Recognition
- Multimodal Machine Learning Applications
- AI in cancer detection
- Advanced machining processes and optimization
- Advanced Neural Network Applications
- Advanced Computing and Algorithms
- Advanced Condensed Matter Physics
- Machine Fault Diagnosis Techniques
- Radiation Dose and Imaging
- Medical Image Segmentation Techniques
- Machine Learning and ELM
- Multiferroics and related materials
- Magnetic and transport properties of perovskites and related materials
- Advanced Algorithms and Applications
- Head and Neck Cancer Studies
- Gear and Bearing Dynamics Analysis
- Face recognition and analysis
Sichuan University
2024-2025
Chengdu University
2024
Chongqing University
2024
University of Chinese Academy of Sciences
2018-2022
Institute of Computing Technology
2022
Chinese Academy of Sciences
2022
To obtain high-quality positron emission tomography (PET) images while minimizing radiation exposure, numerous methods have been dedicated to acquiring standard-count PET (SPET) from low-count (LPET). However, current failed take full advantage of the different emphasized information multiple domains, i.e., sinogram, image, and frequency resulting in loss crucial details. Meanwhile, they overlook unique inner-structure sinograms, thereby failing fully capture its structural characteristics...
Semi-supervised learning (SSL) has shown notable potential in relieving the heavy demand of dense prediction tasks on large-scale well-annotated datasets, especially for challenging multi-organ segmentation (MoS). However, prevailing class-imbalance problem MoS caused by substantial variations organ size exacerbates difficulty SSL network. To address this issue, paper, we propose an innovative semi-supervised network with BAlanced Subclass regularIzation and semantic-Conflict penalty...
Multi-Source Domain Adaptation (MSDA) aims at transferring knowledge from multiple labeled source domains to benefit the task in an unlabeled target domain. The challenges of MSDA lie mitigating domain gaps and combining information diverse domains. In most existing methods, can be jointly or separately aligned this work, we consider that these two types i.e. joint separate alignments, are complementary propose a mutual learning based alignment network (MLAN) combine their advantages....
Universal Multi-source Domain Adaptation (UniMDA) transfers knowledge from multiple labeled source domains to an unlabeled target domain under shifts (different data distribution) and class (unknown classes). Existing solutions focus on excavating image features detect unknown samples, ignoring abundant information contained in textual semantics. In this paper, we propose Adaptive Prompt learning with Negative semantics uncErtainty modeling method based Contrastive Language-Image...
To acquire high-quality positron emission tomography (PET) images while reducing the radiation tracer dose, numerous efforts have been devoted to reconstructing standard-dose PET (SPET) from low-dose (LPET). However, success of current fully-supervised approaches relies on abundant paired LPET and SPET images, which are often unavailable in clinic. Moreover, these methods mix dose-invariant content with dose level-related dose-specific details during reconstruction, resulting distorted...
Organ delineation is critical for diagnosis and treatment planning so as to attract a lot of attention. Recently, neural network based methods yield accurate segmentation metrics like dice coefficient. However, they have face the problem indistinct boundaries since usually modeled pixel classification task ignoring anatomical priors. Inspired by fact that information an essential prior doctors in organ segmentation, this paper proposes mesh regression-based shape enhancement operator. This...