A5014680608- Retinal Imaging and Analysis
- Intracranial Aneurysms: Treatment and Complications
- Cerebrovascular and Carotid Artery Diseases
- Retinal and Optic Conditions
- Optical Coherence Tomography Applications
- Digital Imaging for Blood Diseases
- Aortic aneurysm repair treatments
- Glaucoma and retinal disorders
- Acute Ischemic Stroke Management
- Vascular Malformations Diagnosis and Treatment
- Retinal Diseases and Treatments
- Medical Image Segmentation Techniques
- Aortic Disease and Treatment Approaches
- Intracerebral and Subarachnoid Hemorrhage Research
- Traumatic Brain Injury and Neurovascular Disturbances
- MRI in cancer diagnosis
- Cardiac Valve Diseases and Treatments
- Electrospun Nanofibers in Biomedical Applications
- Renal cell carcinoma treatment
- RNA Interference and Gene Delivery
- Retinal and Macular Surgery
- Topic Modeling
- Sentiment Analysis and Opinion Mining
- Tissue Engineering and Regenerative Medicine
- Cardiovascular Health and Risk Factors
Zhujiang Hospital
2020-2025
Southern Medical University
2020-2025
Guangdong Provincial People's Hospital
2023-2024
Guangdong Academy of Medical Sciences
2023-2024
Macquarie University
2020-2022
First People's Hospital of Foshan
2022
Vision Medicals (China)
2022
Hong Kong University of Science and Technology
2016-2017
University of Hong Kong
2016-2017
Background: Assessment of cerebral aneurysm rupture risk is an important task, but it remains challenging. Recent works applying machine learning to evaluation presented positive results. Yet they were based on limited aspects data, and lack interpretability may limit their use in clinical setting. We aimed develop interpretable models multidimensional data for assessment. Methods: Three hundred seventy-four aneurysms included the study. Demographic, medical history, lifestyle behaviors,...
Skin cancer is one of the most common types cancer. An accessible tool to public can help screening for malign lesion. We aimed develop a deep learning model classify skin lesion using clinical images and meta information collected from smartphones. A neural network was developed with two encoders extracting image data metadata. multimodal fusion module intra-modality self-attention inter-modality cross-attention proposed effectively combine features features. The trained on tested dataset...
Intracranial aneurysms (IAs) are a life-threatening disease. Their rupture can lead to hemorrhagic stroke. Most studies applying deep learning for the detection of based on angiographic images. However, critical diagnostic information such as morphology and aneurysm location not captured by algorithms still require manual assessments.Digital subtraction angiography (DSA) is gold standard diagnosis. To facilitate fully automatic diagnosis aneurysms, we proposed comprehensive system detection,...
Background Aneurysm wall enhancement (AWE) on vessel imaging (VWI) scans is a robust biomarker for aneurysmal vulnerability. This study aimed to explore the association of different sleep patterns with AWE and other vulnerability features. Methods Patients unruptured intracranial aneurysms were prospectively recruited. Sleep characteristics collected through standard questionnaire. Poor quality was defined using Pittsburgh Quality Index (PSQI)>5. Cross-sectional multistate predictive...
Background Intracranial aneurysm wall degradation can be associated with lipid infiltration. However, the relationship between infiltration and rupture has not been explored quantitatively. To investigate correlation rupture, we utilized patient-specific simulation of LDL transport to analyze in cerebral wall. Methods Sixty two aneurysms were analyzed. Patient blood pressure, plasma concentration three dimensional angiographic images obtained simulate aneurysms. Morphological, hemodynamic...
Purpose: This study aimed to develop artificial intelligence models for predicting postoperative functional outcomes in patients with rhegmatogenous retinal detachment (RRD). Methods: A retrospective review and data extraction were conducted on 184 diagnosed RRD who underwent pars plana vitrectomy (PPV) gas tamponade. The primary outcome was the best-corrected visual acuity (BCVA) at three months after surgery. Those a BCVA of less than 6/18 Snellen classified into vision impairment group....
Abstract BACKGROUND The morphological and hemodynamic features differ between middle cerebral artery (MCA) bifurcations with without aneurysms. OBJECTIVE To investigate the differences aneurysmal MCA bifurcation contralateral nonaneurysmal anatomy. METHODS Computed tomography angiography of 36 patients unilateral small saccular aneurysms was evaluated. parent–daughter angles (φ1 for larger branch φ2 smaller branch), angle (φ = φ1 + φ2), inclination (γ angle), their relationships locations...
We present a coronary vessel segmentation method for X-Ray angiography images using multiresolution and multiscale deep learning. Our constructs set of from an input image via bilinear interpolation, which can handle vessels with uneven distribution contrast. incorporate Multiresolution Multiscale Convolution Filtering into U-Net Network, help to improve accuracy results by dealing various thickness in different positions. investigate two types experiments strategy U-Net, respectively. has...
Background: The prediction of aneurysm treatment outcomes can help to optimize the strategies. Machine learning (ML) has shown positive results in many clinical areas. However, development such models requires expertise ML, which is not an easy task for surgeons. Objectives: recently emerged automated machine (AutoML) promise making ML more accessible non-computer experts. We aimed evaluate feasibility applying AutoML develop outcome prediction. Methods: patients with aneurysms treated by...
Rapid endothelialization is extremely essential for the success of small-diameter tissue-engineered vascular graft (TEVG) (<6 mm) transplantation. However, severe inflammation in situ often causes cellular energy decline endothelial cells. The supply involved therapy remains unclear, and whether promoting would be helpful regeneration grafts needs to established. In our work, we generated an AMPK activator (5-aminoimidazole-4-carboxamide ribonucleotide, AICAR) immobilized graft....
Specifying generic flow boundary conditions in aneurysm hemodynamic simulations yields a great degree of uncertainty for the evaluation rupture risk. Herein, we proposed use flowrate-independent parameters discriminating unstable aneurysms and compared their prognostic performance against that conventional absolute parameters. This retrospective study included 186 collected from three international centers, with stable having minimum follow-up period 24 months. The aneurysmal wall shear...
Conventional OCT retinal disease classification methods primarily rely on fully supervised learning, which requires a large number of labeled images. However, sometimes the images in private domain is small but there exists annotated open dataset public domain. In response to this scenario, new transfer learning method based sub-domain adaptation (TLSDA), involves first and then fine-tuning, was proposed study. Firstly, modified deep network with pseudo-label (DSAN-PL) align feature spaces...