A5032572295- Retinal Imaging and Analysis
- Corneal Surgery and Treatments
- Digital Imaging for Blood Diseases
- Glaucoma and retinal disorders
- Radiomics and Machine Learning in Medical Imaging
- Cardiovascular Function and Risk Factors
- Cardiac Valve Diseases and Treatments
- Meteorological Phenomena and Simulations
- Cardiac, Anesthesia and Surgical Outcomes
- Computer Graphics and Visualization Techniques
- Cardiac and Coronary Surgery Techniques
- Aerospace and Aviation Technology
- Cardiac Imaging and Diagnostics
- Lung Cancer Diagnosis and Treatment
- Advanced Vision and Imaging
- COVID-19 diagnosis using AI
- Generative Adversarial Networks and Image Synthesis
- Target Tracking and Data Fusion in Sensor Networks
- Retinal Diseases and Treatments
- Advanced X-ray and CT Imaging
Peking Union Medical College Hospital
2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2024
Huzhou University
2021-2024
Shaoxing University
2024
ShanghaiTech University
2023
China Aerodynamics Research and Development Center
2021
Purpose: The discrepancy of the number between ophthalmologists and patients in China is large. Retinal vein occlusion (RVO), high myopia, glaucoma, diabetic retinopathy (DR) are common fundus diseases. Therefore, this study, a five-category intelligent auxiliary diagnosis model for diseases proposed; model's area focus marked. Methods: A total 2000 images were collected; 3 different 5-category models trained via transfer learning image preprocessing techniques. 1134 used testing. clinical...
The lack of primary ophthalmologists in China results the inability basic-level hospitals to diagnose pterygium patients. To solve this problem, an intelligent-assisted lightweight diagnosis model based on anterior segment images is proposed study. Pterygium a common and frequently occurring disease ophthalmology, fibrous tissue hyperplasia both diagnostic biomarker surgical biomarker. diagnosed biomarkers pterygium. First, total 436 were collected; then, two models (MobileNet 1 MobileNet 2)...
Objective: This study aims to implement and investigate the application of a special intelligent diagnostic system based on deep learning in diagnosis pterygium using anterior segment photographs. Methods: A total 1,220 photographs normal eyes patients were collected for training (using 750 images) testing 470 develop an model. The images classified into three categories by experts system: (i) group, (ii) observation group pterygium, (iii) operation pterygium. results compared with those...
A two-category model and a segmentation of pterygium were proposed to assist ophthalmologists in establishing the diagnosis ophthalmic diseases. total 367 normal anterior segment images collected at Affiliated Eye Hospital Nanjing Medical University. AlexNet, VGG16, ResNet18, ResNet50 models used train models. 150 test models, results compared. The main evaluation indicators, including sensitivity, specificity, area under curve, kappa value, receiver operator characteristic curves four...
Abstract The development of neural relighting techniques has by far outpaced the rate their corresponding training data ( e.g., OLAT) generation. For example, high-quality from a single portrait image still requires supervision comprehensive datasets covering broad diversities in gender, race, complexion, and facial geometry. We present hybrid parametric (PN-Relighting) framework for relighting, using much smaller OLAT dataset or SMOLAT. At core PN-Relighting, we employ 3D faces coupled with...
This paper proposes a new algorithm for the aerodynamic parameter and noise estimation aircraft dynamical systems. The Bayesian inference method is combined with an unscented Kalman filter to estimate augmented states unknown covariance parameters jointly. A Gauss‐Newton utilized sequentially maximize posterior likelihood function estimation. performance of proposed evaluated compared two other UKFs via flight scenario given aircraft. results indicate that has equivalent simplified UKF prior...
AIM: To evaluate the application of an intelligent diagnostic model for pterygium. METHODS: For diagnosis pterygium, attention mechanisms—SENet, ECANet, CBAM, and Self-Attention—were fused with lightweight MobileNetV2 structure to construct a tri-classification model. The study used 1220 images three types anterior ocular segments pterygium provided by Eye Hospital Nanjing Medical University. Conventional classification models—VGG16, ResNet50, MobileNetV2, EfficientNetB7—were trained on same...
Segmenting the left ventricle from transgastric short-axis views (TSVs) on transesophageal echocardiography (TEE) is cornerstone for cardiovascular assessment during perioperative management. Even seasoned professionals, procedure remains time-consuming and experience-dependent. The current study aims to evaluate feasibility of deep learning automatic segmentation by assessing validity different U-Net algorithms. A large dataset containing 1388 TSV acquisitions was retrospectively collected...
In cardiac surgical procedures, patients carrying high-risk profiles are prone to encompass complicated cardiopulmonary bypass (CPB) separation. Intraoperative transesophageal echocardiography (TEE), a readily available tool, is utilized detect structural and functional pathologies as well facilitate clinical management of CPB separation, especially in the episodes hemodynamic compromise. However, conventional TEE examination, always performed liberal fashion without any restriction view...
The automatic left ventricle segmentation in transesophageal echocardiography (TEE) is of significant importance. In this paper, we constructed a large-scale TEE apical four-chamber view (A4CV) image dataset and proposed an ventricular method for the A4CV based on UNeXt deep neural network.
The detection of thoracic abnormalities challenge is organized by the Deepwise AI Lab. divided into two rounds. In this paper, we present results 6 teams which reach second round. adopts ChestX-Det10 dateset proposed first chest X-Ray dataset with instance-level annotations, including 10 categories disease/abnormality 3,543 images. annotations are located at https://github.com/Deepwise-AILab/ChestX-Det10-Dataset. challenge, randomly split all data 3001 images for training and 542 testing.