A5101650644- Advanced Radiotherapy Techniques
- Medical Imaging Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Advanced X-ray and CT Imaging
- Radiation Therapy and Dosimetry
- Meat and Animal Product Quality
- Management of metastatic bone disease
- Lung Cancer Diagnosis and Treatment
- Radiation Dose and Imaging
- Polysaccharides Composition and Applications
- Advanced Image Processing Techniques
- Electric Motor Design and Analysis
- Advanced Measurement and Detection Methods
- AI in cancer detection
- Generative Adversarial Networks and Image Synthesis
- Augmented Reality Applications
- Environmental and Agricultural Sciences
- Advanced Neural Network Applications
- Photoacoustic and Ultrasonic Imaging
- Particle accelerators and beam dynamics
- Multilevel Inverters and Converters
- Food Supply Chain Traceability
- Remote-Sensing Image Classification
- Image and Video Stabilization
- Surgical Simulation and Training
Nanjing Medical University
2018-2025
Changzhou No.2 People's Hospital
2018-2025
Zhejiang Ocean University
2023-2025
Jiangsu University
2019-2023
Second Affiliated Hospital of Nanjing Medical University
2020
Weatherford College
2020
Beijing Institute of Graphic Communication
2017
Shanghai Jiao Tong University
2015
Ministry of Education of the People's Republic of China
2015
To develop high-quality synthetic CT (sCT) generation method from low-dose cone-beam (CBCT) images by using attention-guided generative adversarial networks (AGGAN) and apply these to dose calculations in radiotherapy. The CBCT/planning of 170 patients undergoing thoracic radiotherapy were used for training testing. CBCT scanned under a fast protocol with 50% less clinical projection frames compared standard chest M20 protocol. Training aligned paired was performed conditional (so-called...
To propose a synthesis method of pseudo-CT (CTCycleGAN) images based on an improved 3D cycle generative adversarial network (CycleGAN) to solve the limitations cone-beam CT (CBCT), which cannot be directly applied correction radiotherapy plans.The U-Net with residual connection and attention gates was used as generator, discriminator full convolutional neural (FCN). The imaging quality is by adding gradient loss function. Fivefold cross-validation performed validate our model. Each pseudo...
An increasingly popular non-thermal approach for food preservation is atmospheric cold plasma (ACP), which can efficiently inactivate various microorganisms. This study explored the impact of different ACP treatment modes on red shrimp in chain storage. A novel mode, cyclical, was introduced as an alternative to one-time treatment. The results indicated that cyclical extended ozone and exposure time 35 min compared treatment, resulting a notable enhancement effectiveness. While significantly...
Cone-beam computed tomography (CBCT) is widely used for daily image guidance in radiation therapy, enhancing the reproducibility of patient setup. However, its application adaptive radiotherapy (ART) limited by many imaging artifacts and inaccurate Hounsfield units (HUs). The correction CBCT necessary great value CBCT-based ART.To explore synthetic CT (sCT) generation from images thorax abdomen patients, which usually surfer serious duo to organ state changes. In this study, a streaking...
Abstract Background The delineation of organs at risk (OARs) and clinical target volume (CTV) is an important step in adaptive radiotherapy (ART). Cone‐beam computed tomography (CBCT) images are easy to obtain (RT). Objectives This study aims develop effective CBCT‐guided method for breast cancer ART. Methods A total 60 planning CT 330 CBCT from patients with who underwent breast‐conserving surgery were used uncertainty‐guided multitask semi‐supervised network (UGMNet), which guided by model...
Introduction Multiple targets with varying distances are common in radiotherapy. Reducing treatment time the plan design helps minimize patient movement and discomfort during process. This retrospective study aimed to investigate impact of intertarget (ITDs) on dosimetric differences delivery efficiency two single-isocenter techniques. Methods ITDs for 15 patients dual-site vertebral metastases undergoing volume-modulated arc therapy (VMAT) were modified using Matlab 2019a. Distances 2, 4,...
Diffuse large B-cell lymphoma (DLBCL), a cancer of B cells, has been one the most challenging and complicated diseases because its considerable variation in clinical behavior, response to therapy, prognosis. Radiomic features from medical images, such as PET have become valuable for disease classification or prognosis prediction using learning-based methods. In this paper, new flexible ensemble deep learning model is proposed DLBCL 18F-FDG images. This study proposes multi-R-signature...
Objective.A multi-discriminator-based cycle generative adversarial network (MD-CycleGAN) model is proposed to synthesize higher-quality pseudo-CT from MRI images.Approach.MRI and CT images obtained at the simulation stage with cervical cancer were selected train model. The generator adopted DenseNet as main architecture. local global discriminators based on a convolutional neural jointly discriminated authenticity of input image data. In testing phase, was verified by fourfold...
In modern radiotherapy, error reduction in the patients' daily setup is important for achieving accuracy. our study, we proposed a new approach development of an assist system radiotherapy position by using augmented reality (AR). We aimed to improve accuracy patients undergoing and evaluate who were diagnosed with head neck cancer, that chest abdomen cancer. acquired patient's simulation CT data three-dimensional (3D) reconstruction external surface organs. The AR tracking software detected...
This study aimed to present a deep-learning network called contrastive learning-based cycle generative adversarial networks (CLCGAN) mitigate streak artifacts and correct the CT value in four-dimensional cone beam computed tomography (4D-CBCT) for dose calculation lung cancer patients.
Magnetic resonance imaging (MRI) plays an important role in clinical diagnosis, but it is susceptible to metal artifacts. The generative adversarial network GatedConv with gated convolution (GC) and contextual attention (CA) was used inpaint the artifact region MRI images.MRI images containing or near teeth of 70 patients were collected, scanning sequence a T1-weighted high-resolution isotropic volume examination sequence. A total 10 000 slices obtained after data enhancement, which 8000 for...
Cone-beam computed tomography (CBCT) is widely used in clinical radiotherapy, but its small field of view (sFOV) limits application potential. In this study, a transformer-based dual-domain network (dual_swin), which combined image domain restoration and sinogram restoration, was proposed for the reconstruction complete CBCT images with extended FOV from truncated sinograms.The planning CT large (LFOV) 330 patients who received radiation therapy were collected. The synthetic (sCBCT) LFOV...
Abstract Background and objective Metallic magnetic resonance imaging (MRI) implants can introduce field distortions, resulting in image distortion, such as bulk shifts signal‐loss artifacts. Metal Artifacts Region Inpainting Network (MARINet), using the symmetry of brain MRI images, has been developed to generate normal images domain improve quality. Methods T1‐weighted containing or located near teeth 100 patients were collected. A total 9000 slices obtained after data augmentation. Then,...
Objective: To generate virtual non-contrast (VNC) computed tomography (CT) from intravenous enhanced CT through convolutional neural networks (CNN) and compare calculated dose among CT, VNC, real scanning. Method: 50 patients who accepted scanning before after contrast agent injections were selected, two sets of images registered. A total 40 10 groups used as training test datasets, respectively. The U-Net architecture was applied to learn the relationship between CT. VNC generated in...
Abstract Herein, a Harris corner detection algorithm is proposed based on the concepts of iterated threshold segmentation and adaptive iterative (AIT–Harris), stepwise local stitching used to obtain wide-field ultrasound (US) images. Cone-beam computer tomography (CBCT) US images from 9 cervical cancer patients 1 prostate patient were examined. In experiment, features extracted AIT–Harris, Harris, Morave algorithms. Accordingly, ultrasonic obtained after stitching, matching rates all tested...