A5091066995- Radiomics and Machine Learning in Medical Imaging
- Dental Radiography and Imaging
- Lung Cancer Diagnosis and Treatment
- COVID-19 diagnosis using AI
- AI in cancer detection
- Medical Imaging and Analysis
- Advanced X-ray and CT Imaging
- Petroleum Processing and Analysis
- Medical Image Segmentation Techniques
- Dental Implant Techniques and Outcomes
- Enhanced Oil Recovery Techniques
- Medical Imaging Techniques and Applications
- Hydrocarbon exploration and reservoir analysis
- Orthodontics and Dentofacial Orthopedics
- 3D Shape Modeling and Analysis
- Advanced Image and Video Retrieval Techniques
- Face recognition and analysis
- Anatomy and Medical Technology
- Computer Graphics and Visualization Techniques
- Generative Adversarial Networks and Image Synthesis
- Artificial Intelligence in Healthcare
- Augmented Reality Applications
- Infectious Diseases and Tuberculosis
- Image Retrieval and Classification Techniques
- Text and Document Classification Technologies
Sichuan University
2013-2025
Henan University
2025
Zhongshan People's Hospital
2025
Chengdu University
2016-2024
China University of Petroleum, Beijing
2023-2024
Jiangnan University
2023-2024
State Key Laboratory of Oral Diseases
2016
October 6 University
2016
Stomatology Hospital
2016
Chinese University of Hong Kong
2008-2012
Diabetic retinopathy (DR) is a common eye disease and significant cause of blindness in diabetic patients. Regular screening with fundus photography timely intervention the most effective way to manage disease. The large population patients their massive requirements have generated interest computer-aided fully automatic diagnosis DR. Deep neural networks, on other hand, brought many breakthroughs various tasks recent years. To automate DR provide appropriate suggestions patients, we built...
Abstract Deep learning, transforming input data into target prediction through intricate network structures, has inspired novel exploration in automated diagnosis based on medical images. The distinct morphological characteristics of chest abnormalities between drug‐resistant tuberculosis (DR‐TB) and drug‐sensitive (DS‐TB) computed tomography (CT) are potential value differential diagnosis, which is challenging the clinic. Hence, 1176 CT volumes from equal number patients with (TB), we...
To develop and validate a modified deep learning (DL) model based on nnU-Net for classifying segmenting five-class jaw lesions using cone-beam CT (CBCT).
Lung cancer postoperative complication prediction (PCP) is significant for decreasing the perioperative mortality rate after lung surgery. In this paper we concentrate on two PCP tasks: (1) binary classification predicting whether a patient will have complications; and (2) three-class multi-label which experience. Furthermore, an important clinical requirement of extraction crucial variables from electronic medical records. We propose novel multi-layer perceptron (MLP) model called MLP...
The detection of epidermal growth factor receptor (EGFR) mutation and programmed death ligand-1 (PD-L1) expression status is crucial to determine the treatment strategies for patients with non-small-cell lung cancer (NSCLC). Recently, rapid development radiomics including but not limited deep learning techniques has indicated potential role medical images in diagnosis diseases. Eligible diagnosed/treated at West China Hospital Sichuan University from January 2013 April 2019 were identified...
Existing challenges of lung cancer screening included non-accessibility computed tomography (CT) scanners and inter-reader variability, especially in resource-limited areas. The combination mobile CT deep learning technique has inspired innovations the routine clinical practice. This study recruited participants prospectively two rural sites western China. A system was developed to assist clinicians identify nodules evaluate malignancy with state-of-the-art performance assessed by recall,...
Surgical reconstruction of mandibular defects is a clinical routine manner for the rehabilitation patients with deformities. The mandible plays crucial role in maintaining facial contour and ensuring speech mastication functions. repairing significant yet challenging task oral–maxillofacial surgery. Currently, mainly available methods are traditional digitalized design that suffer from substantial artificial operations, limited applicability high error rates. An automated, precise,...
<title>Abstract</title> <bold>Objective:</bold> Three-dimensional (3D) landmark detection is essential for assessing craniofacial growth and planning surgeries such as orthodontic, orthognathic, traumatic, plastic procedures. This study aimed to develop an automatic 3D landmarking model oral maxillofacial regions validate its accuracy, robustness, generalizability in both spiral computed tomography (SCT, 41 landmarks) cone-beam (CBCT, 14 scans. <bold>Methods:</bold> The was constructed using...
Accurate virtual orbital reconstruction is crucial for preoperative planning. Traditional methods, such as the mirroring technique, are unsuitable defects involving both sides of midline and time-consuming labor-intensive. This study introduces a modified 3D U-Net+++ architecture reconstruction, aiming to enhance precision automation. The model was trained tested with 300 synthetic from cranial spiral CT scans. method validated in 15 clinical cases fractures evaluated using quantitative...
Survival analysis is important for guiding further treatment and improving lung cancer prognosis. It a challenging task because of the poor distinguishability features missing values in practice. A novel multi-task based neural network, SurvNet, proposed this paper. The SurvNet model trained learning framework to jointly learn across three related tasks: input reconstruction, survival classification, Cox regression. uses an reconstruction mechanism cooperating with incomplete-aware loss...
X-ray imaging frequently introduces varying degrees of metal artifacts to computed tomography (CT) images when implants are present. For the artifact reduction (MAR) task, existing end-to-end methods often exhibit limited generalization capabilities. While based on multiple iterations suffer from accumulative error, resulting in lower-quality restoration outcomes. In this work, we innovatively present a generalized diffusion model for Metal Artifact Reduction (DiffMAR). The proposed method...
Radiographic attributes of lung nodules remedy the shortcomings cancer computer-assisted diagnosis systems, which provides interpretable diagnostic reference for doctors. However, current studies fail to dedicate multi-label classification using convolutional neural networks (CNNs) and are inferior in exploiting statistical dependency between labels. In addition, data imbalance is an indispensable problem be reckoned with when employing CNNs perform nodule classification. It introduces...