A5051440156- Radiomics and Machine Learning in Medical Imaging
- Atrial Fibrillation Management and Outcomes
- Cardiac electrophysiology and arrhythmias
- Medical Imaging Techniques and Applications
- AI in cancer detection
- Acupuncture Treatment Research Studies
- Cardiac Arrhythmias and Treatments
- Healthcare and Venom Research
- Advanced X-ray and CT Imaging
- Lung Cancer Diagnosis and Treatment
- Medical Image Segmentation Techniques
- Traditional Chinese Medicine Studies
- Advanced Radiotherapy Techniques
- Brain Tumor Detection and Classification
- MRI in cancer diagnosis
- Advanced Image and Video Retrieval Techniques
- Advanced biosensing and bioanalysis techniques
- Advanced Image Fusion Techniques
- Extracellular vesicles in disease
- Medical Imaging and Analysis
- ECG Monitoring and Analysis
- Intracranial Aneurysms: Treatment and Complications
- Privacy-Preserving Technologies in Data
- MicroRNA in disease regulation
- Nanoplatforms for cancer theranostics
Guangxi Medical University
2023-2025
Yonsei University
2023-2024
Tokyo Institute of Technology
2019-2024
Huazhong University of Science and Technology
2024
Yuncheng University
2024
Suzuki (Japan)
2023
Yonsei University Health System
2021-2023
Hebei University
2023
National Institute of Information and Communications Technology
2021-2023
Gangnam Severance Hospital
2023
Exosomes are extracellular vesicles comprising bilayer phospholipid membranes and secreted by eukaryotic cells. They released via cellular exocytosis, contain DNA, RNA, proteins, other substances, participate in various communications between tissues organs. Since the discovery of exosomes 1983, animal-derived have become a research focus for small-molecule drug delivery biology, medicine, fields owing to their good biocompatibility homing effects. Recent studies found that plant-derived...
Dramatic development in perovskite solar cells (PSCs) and the widespread application of wearable electronics have attracted extensive research area large-scale flexible power sources based on PSCs. Manufacturing PSCs by printing is considered to be one most potential methods. However, it still a great challenge print large-area uniform hole transport layers (HTLs) rough soft plastic substrate achieve with high efficiency good stability. Herein, we synthesized viscous poly(3,4-ethylene...
Remote sensing images have the essential attribute of large-scale spatial variation and complex scene information, as well high similarity between various classes significant differences within same class, which are easy to cause misclassification. To solve this problem, an efficient systematic architecture named EFCOMFF-Net (Multi-scale Feature Fusion Network with Enhanced Correlation) is proposed reduce gap among multi-scale features fuse them improve representation ability remote images....
Abstract Although pulmonary vein isolation (PVI) gaps and extrapulmonary triggers contribute to recurrence after atrial fibrillation (AF) ablation, their precise mechanisms remain unproven. Our study assessed the impact of PVI on rhythm outcomes using a human AF digital twin. We included 50 patients (76.0% with persistent AF) who underwent catheter ablation realistic twin by integrating computed tomography electroanatomical mapping. evaluated final status, including tachycardia (AT), across...
We have proposed a computer-assisted framework for machine-learning-based delineation of gross tumor volumes (GTVs) following an optimum contour selection (OCS) method. The key idea the was to feed image features around GTV contours (determined based on knowledge radiation oncologists) into machine-learning classifier during training step, after which produces 'degree GTV' each voxel in testing step. Initial regions were extracted using support vector machine (SVM) that learned inside and...
Purpose: Various kinds of enhancement filters have been developed in computer‐aided diagnostic (CAD) frameworks for asymptomatic intracranial aneurysms magnetic resonance angiography (MRA). However, many bending or branching portions on vessels are also enhanced by the conventional as false positives 3.0 T MRA, which can visualize smaller compared with 1.5 MRA. To overcome this problem, study focused developing an ellipsoid convex (ECE) filter, selectively enhance while reducing positive...
Abstract Background Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common worldwide and few drugs are available for its treatment. Lycorine has effective anti-inflammatory lipid-lowering effects, but impact on MASLD not fully understood. In this study, we intend to test intervention effect of lycorine MASLD. Methods A mouse model was constructed a high-fat diet 16 weeks, low, medium, high doses were given by gavage last 4 weeks. Detecting indicators related...
To assist radiation oncologists in the delineation of tumor regions during treatment planning for lung cancer, we have proposed an automated contouring algorithm based on optimum contour selection (OCS) method computed tomography (CT) images with positron emission (PET)/CT images. The basic concept OCS is to select a global object multiple active delineations level set around tumors. First, PET were registered CT by using affine transformation matrices. initial gross volume (GTV) each was...
Objective.In clinical medicine, localization and identification of disease on spinal radiographs are difficult require a high level expertise in the radiological discipline extensive experience. The model based deep learning acquires certain recognition abilities through continuous training, thereby assisting physicians diagnosis. This study aims to develop an object detection network that accurately locates classifies abnormal parts x-ray photographs.Approach.This proposes learning-based...
Background Although pulmonary vein isolation (PVI) gaps contribute to recurrence after atrial fibrillation (AF) catheter ablation, the mechanism is unclear. We used realistic computational human AF modeling explore wave-dynamic changes of PVI with (PVI-gaps). Methods included 40 patients (80% male, 61.0 ± 9.8 years old, 92.5% persistent AF) who underwent ablation develop our model. compared effects a complete (CPVI) and PVI-gap (2-mm × 4) on wave-dynamics by evaluating dominant frequency...
We have developed an automated method for extraction of lung tumors using a machine learning classifier with knowledge radiation oncologists on data sets treatment planning computed tomography (CT) and 18F-fluorodeoxyglucose (FDG)-positron emission (PET)/CT images. First, the PET images were registered CT through diagnostic PET/CT. Second, six voxel-based features including voxel values magnitudes image gradient vectors derived from each in /CT sets. Finally, extracted by support vector...
Background: The efficacy of antiarrhythmic drugs (AAD) can vary in patients with atrial fibrillation (AF), and the PITX2 gene affects responsiveness AADs. We explored virtual AAD (V-AAD) responses between wild-type PITX2+/--deficient AF conditions by realistic silico modeling. Methods: tested V-AADs modeling integrated patients' 3D-computed tomography 3D-electroanatomical mapping, acquired 25 (68% male, 59.8 ± 9.8 years old, 32.0% paroxysmal type). ion currents for PITX2+/- deficiency each...
Federated Learning (FL) enables multiple institutes to train models collaboratively without sharing private data. Most of the current FL research focuses on perspectives such as communication efficiency, privacy protection, and personalization. Almost all work assumed that data are already ideally collected. However, in medical image analysis scenarios, annotation demands both expertise tedious labor, which means it is a critical problem cannot be neglected FL. In this study, we proposed...
Background: We previously reported that a computational modeling-guided antiarrhythmic drug (AAD) test was feasible for evaluating multiple AADs in patients with atrial fibrillation (AF). explored the anti-AF mechanisms of and spatial change AF wave-dynamics by realistic model. Methods: used modeling 25 (68% male, 59.8 ± 9.8 years old, 32.0% paroxysmal AF) reflecting anatomy, histology, electrophysiology left atrium (LA) to characterize effects five (amiodarone, sotalol, dronedarone,...
Deep learning requires a large dataset for training, but collecting and annotating such are time-consuming often very difficult. In this study, we proposed 3D massive-training artificial neural network (MTANN) incorporated with Hessian-based enhancer to achieve high performance of our MTANN model small training datasets. Contrast-enhanced CT scans 42 patients 194 liver tumors from the Liver Tumor Segmentation (LiTS) Benchmark database were used in study. models trained with: 14 28 tumors....
The lack of systematic classification and standard treatment principles for knee ankylosis prevents optimal treatments. This study explored treatments type I (mild) joint ankylosis. retrospective analysed patients with admitted from March 2013 to January 2018 who underwent sequential arthroscopic release. 62 had 12-36 (average, 18) months follow-up. Thirty-eight were released; these, 18 assisted by limited incision partial quadriceps femoris expansion myotomy released according arthroscopy....
In this study, we proposed a deep-learning-based semantic segmentation of breast tumors in diagnostic MRI, which had been planned with chemotherapy treatment, for extracting imaging biomarkers radiomics/radiogenomics studies. Deep-learning neural network convolution (NNC) that employs patched-based regression convolutional manner were employed to output map the likelihood being tumor. We trained our NNC three dynamic contrast-enhanced MR images as input: pre-contrast, early-phase, and...