A5081689932- Advanced MRI Techniques and Applications
- Advanced Neuroimaging Techniques and Applications
- Cardiac Imaging and Diagnostics
- Medical Imaging Techniques and Applications
- Advanced X-ray and CT Imaging
- Brain Tumor Detection and Classification
- Pancreatic and Hepatic Oncology Research
- Radiation Dose and Imaging
- Traumatic Brain Injury and Neurovascular Disturbances
- Cardiovascular Function and Risk Factors
- Traumatic Brain Injury Research
- Coronary Interventions and Diagnostics
- Radiomics and Machine Learning in Medical Imaging
- Cardiac Valve Diseases and Treatments
- MRI in cancer diagnosis
- Multiple Sclerosis Research Studies
- AI in cancer detection
- Neuroendocrine Tumor Research Advances
- Medical Imaging and Analysis
- Cerebral Palsy and Movement Disorders
- Advanced Radiotherapy Techniques
- Optical Coherence Tomography Applications
- Medical Image Segmentation Techniques
- Retinal Imaging and Analysis
- Glaucoma and retinal disorders
University of Malaya
2016-2025
Tongji Hospital
2024
Huazhong University of Science and Technology
2024
Beijing Technology and Business University
2024
Taihe Hospital
2023
Hubei University of Medicine
2023
Peking Union Medical College Hospital
2021-2022
Chinese Academy of Medical Sciences & Peking Union Medical College
2021-2022
National Heart Institute
2018
University Malaya Medical Centre
2006-2013
PurposeWe present the implementation of e-learning in Master Medical Physics programme at University Malaya during a partial lockdown from March to June 2020 due COVID-19 pandemic.MethodsTeaching and Learning (T&L) activities were conducted virtually on platforms. The students' experience feedback evaluated after 15 weeks.ResultsWe found that while students preferred face-to-face, physical teaching, they able adapt new norm e-learning. More than 60% agreed pre-recorded lectures viewing...
Background Left ventricle (LV) structure and functions are the primary assessment performed in most clinical cardiac MRI protocols. Fully automated LV segmentation might improve efficiency reproducibility of assessment. Purpose To develop validate a fully neural network regression‐based algorithm for MRI, with full coverage from apex to base across all phases, utilizing both short axis (SA) long (LA) scans. Study Type Cross‐sectional survey; diagnostic accuracy. Subjects In all, 200 subjects...
Intravascular optical coherence tomography (OCT) is an imaging modality commonly used in the assessment of coronary artery diseases during percutaneous intervention. Manual segmentation to assess luminal stenosis from OCT pullback scans challenging and time consuming. We propose a linear-regression convolutional neural network automatically perform vessel lumen segmentation, parameterized terms radial distances catheter centroid polar space. Benchmarked against gold-standard manual our...
The association between body mass index (BMI) and cognitive impairment (CI) has been the subject of extensive research, yet precise dose-response effects remain undefined.
Tumor-related epilepsy is a prevalent condition in patients with gliomas. Accurate prediction of crucial for early treatment. This study aimed to evaluate the novel application eXtreme Gradient Boost (XGBoost) machine learning (ML) algorithm into radiomics model predicting preoperative tumor-related (PTRE). Its performance was compared 4 conventional ML algorithms, including least absolute shrinkage and selection operator (LASSO), elastic net, random forest, support vector machine. used four...
Purpose: Pancreatic cancer (PACA) is one of the most fatal malignancies worldwide. Immunotherapy largely ineffective in patients with PACA. T-cell exhaustion contributes to immunotherapy resistance. We investigated prognostic potential exhaustion-related genes (TEXGs). Methods: A single-cell RNA (scRNA) sequencing dataset from Tumor Immune Single-Cell Hub (TISCH) and bulk datasets Cancer Genome Atlas (TCGA) Genotype-Tissue Expression (GTEx) were used screen differentially expressed TEXGs....
Cardiac MRI is important for the diagnosis and assessment of various cardiovascular diseases. Automated segmentation left ventricular (LV) endocardium at end-diastole (ED) end-systole (ES) enables automated quantification clinical parameters including ejection fraction. Neural networks have been used general image segmentation, usually via per-pixel categorization e.g. "foreground" "background". In this paper we propose that generally circular LV can be parameterized endocardial contour...
Cine MRI is a clinical reference standard for the quantitative assessment of cardiac function, but reproducibility confounded by motion artefacts. We explore feasibility corrected 3D left ventricle (LV) quantification method, incorporating multislice image registration into model reconstruction, to improve LV functional quantification. Multi-breath-hold short-axis and radial long-axis images were acquired from 10 patients healthy subjects. The proposed framework reduced misalignment between...
The advancement of event cameras has sparked a revolution in imaging technology, presenting exciting opportunities for vision-based measurement tasks. Event operate on an innovative asynchronous principle, which offers several advantages over traditional cameras, including ultra-high dynamic range and exceptional temporal resolution. As result, excel challenging environments characterized by motion blur, overexposure, or underexposure, outperforming conventional frame-based cameras. However,...
To evaluate a 2D-4D registration-cum-segmentation framework for the delineation of left ventricle (LV) in late gadolinium enhanced (LGE) MRI and localization infarcts patient-specific 3D LV models.A 3-step was proposed, consisting of: (1) model reconstruction from motion-corrected 4D cine-MRI; (2) Registration 2D LGE-MRI with (3) contour extraction intersection LGE slices model. The evaluated against cardiac data 27 patients scanned within 6 months after acute myocardial infarction. We...
Abstract Objective . To develop an algorithm to measure slice thickness running on three types of Catphan phantoms with the ability adapt any misalignment and rotation phantoms. Method Images 500, 504, 604 were examined. In addition, images various thicknesses ranging from 1.5 10.0 mm, distance iso-center phantom rotations also The automatic was carried out by processing only objects within a circle having diameter half phantom. A segmentation performed inner dynamic thresholds produce...