A5074311923- Radiomics and Machine Learning in Medical Imaging
- Head and Neck Cancer Studies
- Multimodal Machine Learning Applications
- Retinal Imaging and Analysis
- Surgical Simulation and Training
- COVID-19 diagnosis using AI
- Medical Image Segmentation Techniques
- Retinal Diseases and Treatments
- Advanced Radiotherapy Techniques
- Glaucoma and retinal disorders
- AI in cancer detection
- Medical Imaging and Analysis
- Lung Cancer Diagnosis and Treatment
- Video Surveillance and Tracking Methods
- Topic Modeling
- Intravenous Infusion Technology and Safety
- Neutrophil, Myeloperoxidase and Oxidative Mechanisms
- Viral gastroenteritis research and epidemiology
- Software System Performance and Reliability
- Extracellular vesicles in disease
- Cutaneous Melanoma Detection and Management
- Advanced Steganography and Watermarking Techniques
- Nonmelanoma Skin Cancer Studies
- Antimicrobial Resistance in Staphylococcus
- SARS-CoV-2 detection and testing
Hong Kong University of Science and Technology
2023-2025
University of Hong Kong
2023-2025
Soochow University
2023-2025
Second Affiliated Hospital of Soochow University
2023-2025
Hong Kong Polytechnic University
2024
Shenzhen Institutes of Advanced Technology
2024
University of Electronic Science and Technology of China
2022-2023
Quzhou University
2022
Nasopharyngeal carcinoma (NPC) is a prevalent and clinically significant malignancy that predominantly impacts the head neck area. Precise delineation of Gross Tumor Volume (GTV) plays pivotal role in ensuring effective radiotherapy for NPC. Despite recent methods have achieved promising results on GTV segmentation, they are still limited by lacking carefully-annotated data hard-to-access from multiple hospitals clinical practice. Although some unsupervised domain adaptation (UDA) has been...
Surgical instrument segmentation is fundamentally important for facilitating cognitive intelligence in robot-assisted surgery. Although existing methods have achieved accurate results, they simultaneously generate masks of all instruments, which lack the capability to specify a target object and allow an interactive experience. This paper focuses on novel essential task robotic surgery, i.e., Referring Video Instrument Segmentation (RSVIS), aims automatically identify segment surgical...
Delineation of the clinical target volume (CTV) and organs-at-risk (OARs) is important in cervical cancer radiotherapy. But it generally labor-intensive, time-consuming, subjective. This paper proposes a parallel-path attention fusion network (PPAF-net) to overcome these disadvantages delineation task.The PPAF-net utilizes both texture structure information CTV OARs by employing U-Net capture high-level information, an up-sampling down-sampling (USDS) low-level accentuate boundaries OARs....
Traditional shadow detectors often identify all regions of static images or video sequences. This work presents the Referring Video Shadow Detection (RVSD), which is an innovative task that rejuvenates classic paradigm by facilitating segmentation particular shadows in videos based on descriptive natural language prompts. novel RVSD not only achieves arbitrary areas interest descriptions (flexibility) but also allows users to interact with visual content more directly and naturally using...
Video Shadow Detection (VSD) aims to detect the shadow masks with frame sequence. Existing works suffer from inefficient temporal learning. Moreover, few address VSD problem by considering characteristic (i.e., boundary) of shadow. Motivated this, we propose a Timeline and Boundary Guided Diffusion (TBGDiff) network for where take account past-future guidance boundary information jointly. In detail, design Dual Scale Aggregation (DSA) module better understanding rethinking affinity long-term...
Accurate diabetic retinopathy (DR) grading is crucial for making the proper treatment plan to reduce damage caused by vision loss. This task challenging due fact that DR related lesions are often small and subtle in visual differences intra-class variations. Moreover, relationships between levels complicated. Although many deep learning (DL) systems have been developed with some success, there still rooms accuracy improvement. A common issue not much medical knowledge was used these DL...
Combined Pulmonary Fibrosis and Emphysema (CPFE), recognized as a distinct pulmonary syndrome in 2022, is characterized by unique clinical features pathogenesis that can lead to respiratory failure death. However, the diagnosis of CPFE presents significant challenges impede effective treatment. Here, we assembled multicenter dataset three-dimensional (3D) computed tomography (CT) images patients' lungs from multiple hospitals across different provinces China, including Xiangya Hospital, West...
Expert surgeons often have heavy workloads and cannot promptly respond to queries from medical students junior doctors about surgical procedures. Thus, research on Visual Question Localized-Answering in Surgery (Surgical-VQLA) is essential assist understanding scenarios. Surgical-VQLA aims generate accurate answers locate relevant areas the scene, requiring models identify understand instruments, operative organs, A key issue model's ability accurately distinguish instruments. Current rely...
Respiratory tract infections (RTIs) caused by various pathogens, including viruses, bacteria, and fungi, pose significant public health challenges worldwide. Understanding the etiology epidemiology of RTIs is necessary for clinical management, rational drug use, formulation preventive measures, vaccine development. Quantitative real-time PCR was used to detect analyze respiratory pathogens in outpatients at a hospital Suzhou, FluA, FluB, RSV, ADV, HRV, MP, SARS-CoV-2. Among 27,031 throat...
For robot-assisted surgery, an accurate surgical report reflects clinical operations during surgery and helps document entry tasks, post-operative analysis follow-up treatment. It is a challenging task due to many complex diverse interactions between instruments tissues in the scene. Although existing generation methods based on deep learning have achieved large success, they often ignore interactive relation instrumental tools, thereby degrading performance. This paper presents neural...
Idiopathic pulmonary fibrosis (IPF) is a progressive and fatal lung disease that poses significant challenge to medical professionals due its increasing incidence prevalence coupled with the limited understanding of underlying molecular mechanisms. In this study, we employed novel approach by integrating five expression datasets from bulk tissue single-cell datasets; they underwent pseudotime trajectory analysis, switch gene selection, cell communication analysis. Utilizing prognostic...
Radiation therapy is a primary and effective NasoPharyngeal Carcinoma (NPC) treatment strategy. The precise delineation of Gross Tumor Volumes (GTVs) Organs-At-Risk (OARs) crucial in radiation treatment, directly impacting patient prognosis. Previously, the GTVs OARs was performed by experienced oncologists. Recently, deep learning has achieved promising results many medical image segmentation tasks. However, for NPC segmentation, few public datasets are available model development...
Diabetic retinopathy (DR) is the leading cause of permanent blindness in working-age population, which one common complications diabetes. DR grading crucial determining relevant treatment to reduce vision loss. Automatic approaches are very significant for helping ophthalmologists design adequate patients. However, challenging due facts intra-class variations and inter-class similarities. The key point solving find abundant discriminative lesions corresponding subtle visual differences, such...
Since the outbreak of coronavirus disease 2019 (COVID-19), epidemic has been spreading around world for more than 2 years. Rapid, safe, and on-site detection methods COVID-19 are in urgent demand control epidemic. Here, we established an integrated system, which incorporates a machine-learning-based Fourier transform infrared spectroscopy technique rapid screening air-plasma-based disinfection modules to prevent potential secondary infections. A partial least-squares discrimination analysis...
Robot-assisted surgery has made significant progress, with instrument segmentation being a critical factor in surgical intervention quality. It serves as the building block to facilitate robot navigation and education for next generation of operating intelligence. Although existing methods have achieved accurate results, they simultaneously generate masks all instruments, without capability specify target object allow an interactive experience. This work explores new task Referring Surgical...
Accurate vessel segmentation in Ultra-Wide-Field Scanning Laser Ophthalmoscopy (UWF-SLO) images is crucial for diagnosing retinal diseases. Although recent techniques have shown encouraging outcomes segmentation, models trained on one medical dataset often underperform others due to domain shifts. Meanwhile, manually labeling high-resolution UWF-SLO an extremely challenging, time-consuming and expensive task. In response, this study introduces a pioneering framework that leverages...
Ultra-Wide-Field Scanning Laser Ophthalmoscopy (UWF-SLO) images capture high-resolution views of the retina with typically 200 spanning degrees. Accurate segmentation vessels in UWF-SLO is essential for detecting and diagnosing fundus disease. Recent studies have revealed that selective State Space Model (SSM) Mamba performs well modeling long-range dependencies, which crucial capturing continuity elongated vessel structures. Inspired by this, we propose first Serpentine (Serp-Mamba) network...
Fundus imaging is a pivotal tool in ophthalmology, and different modalities are characterized by their specific advantages. For example, Fluorescein Angiography (FFA) uniquely provides detailed insights into retinal vascular dynamics pathology, surpassing Color Photographs (CFP) detecting microvascular abnormalities perfusion status. However, the conventional invasive FFA involves discomfort risks due to fluorescein dye injection, it meaningful but challenging synthesize images from...
Fundus imaging is a pivotal tool in ophthalmology, and different modalities are characterized by their specific advantages. For example, Fluorescein Angiography (FFA) uniquely provides detailed insights into retinal vascular dynamics pathology, surpassing Color Photographs (CFP) detecting microvascular abnormalities perfusion status. However, the conventional invasive FFA involves discomfort risks due to fluorescein dye injection, it meaningful but challenging synthesize images from...
To develop a deep learning method exploiting active and source-free domain adaptation for gross tumor volume delineation in nasopharyngeal carcinoma (NPC), addressing the variability inaccuracy when deploying segmentation models multicenter multirater settings. One thousand fifty-seven magnetic resonance imaging scans of patients with NPC from 5 hospitals were retrospectively collected annotated by experts same medical group consensus evaluation. data set was used model development (source...
Recent advancements in 3D generation models have opened new possibilities for simulating dynamic object movements and customizing behaviors, yet creating this content remains challenging. Current methods often require manual assignment of precise physical properties simulations or rely on video to predict them, which is computationally intensive. In paper, we rethink the usage multi-modal large language model (MLLM) physics-based simulation, present Sim Anything, a approach that endows...