A5100765644- Radiomics and Machine Learning in Medical Imaging
- Cerebrovascular and Carotid Artery Diseases
- Intracranial Aneurysms: Treatment and Complications
- MRI in cancer diagnosis
- Lung Cancer Diagnosis and Treatment
- Medical Image Segmentation Techniques
- Traumatic Brain Injury and Neurovascular Disturbances
- Brain Tumor Detection and Classification
- Glioma Diagnosis and Treatment
- Pediatric Hepatobiliary Diseases and Treatments
- Organ Transplantation Techniques and Outcomes
- Cardiovascular Health and Disease Prevention
- Lung Cancer Treatments and Mutations
- Advanced Radiotherapy Techniques
- Advanced X-ray and CT Imaging
- Urologic and reproductive health conditions
- 3D Shape Modeling and Analysis
- Sarcoma Diagnosis and Treatment
- Medical Imaging and Analysis
- Diagnosis and treatment of tuberculosis
- Gallbladder and Bile Duct Disorders
- Machine Learning and ELM
- Gaze Tracking and Assistive Technology
- Functional Brain Connectivity Studies
- Parkinson's Disease Mechanisms and Treatments
Chinese Academy of Sciences
2014-2024
Suzhou Institute of Biomedical Engineering and Technology
2016-2024
Fudan University
2020-2022
Huashan Hospital
2020-2021
Army Medical University
2009
Southwest Hospital
2009
Background Differential diagnosis of primary central nervous system lymphoma (PCNSL) and glioblastoma (GBM) is useful to guide treatment strategies. Purpose To investigate the use a convolutional neural network (CNN) model for differentiation PCNSL GBM without tumor delineation. Study Type Retrospective. Population A total 289 patients with (136) or (153) were included, average age cohort was 54 years, there 173 men 116 women. Field Strength/Sequence 3.0 T Axial contrast‐enhanced 1 ‐weighted...
Background Preoperative differentiation of primary central nervous system lymphoma (PCNSL) from glioblastoma (GBM) is important to guide neurosurgical decision‐making. Purpose To validate the generalization ability radiomics models based on multiparametric‐MRI (MP‐MRI) for differentiating PCNSL GBM. Study Type Retrospective. Population In all, 240 patients with GBM ( n = 129) or 111). Field Strength/Sequence 3.0T scanners (two vendors). Sequences: fluid‐attenuation inversion recovery,...
Abstract Background As the rupture of cerebral aneurysm may lead to fatal results, early detection unruptured aneurysms save lives. At present, contrast-unenhanced time-of-flight magnetic resonance angiography is one most commonly used methods for screening aneurysms. The computer-assisted system can help clinicians improve accuracy diagnosis. fully convolutional network could classify image pixel-wise, its three-dimensional implementation highly suitable classification vascular structure....
A new accurate and robust non-rigid point set registration method, named DSMM, is proposed for in the presence of significant amounts missing correspondences outliers. The key idea this algorithm to consider relationship between sets as random variables model prior probabilities via Dirichlet distribution. We assign various each its Student's-t mixture model. later incorporate local spatial representation by representing posterior a linear smoothing filter get closed-form proportions,...
This paper proposes to use multilevel ROI-based features and machine learning method improve the accuracy of qualitative recognition mild cognitive disorder in parkinsonism. 77 Parkinson's patients 32 normal controls with neuropsychological assessments structural magnetic resonance images from Progression Markers Initiative dataset are tested. Specifically, BrainLab software is used process measure volume gray matter, thickness cortex, surface area cortex at each region interest (ROI). We...
Objective This study aimed to evaluate the effectiveness of multi-phase-combined contrast-enhanced CT (CECT) radiomics methods for noninvasive Fuhrman grade prediction clear cell renal carcinoma (ccRCC). Methods A total 187 patients with four-phase CECT images were retrospectively enrolled and then categorized into training cohort (n=126) testing (n=61). All confirmed as ccRCC by histopathological reports. 110 3D classical features extracted from each phase individual lesion, variation also...
Abstract Background Accurate segmentation of unruptured cerebral aneurysms (UCAs) is essential to treatment planning and rupture risk assessment. Currently, three-dimensional time-of-flight magnetic resonance angiography (3D TOF-MRA) has been the most commonly used method for screening due its noninvasiveness. The methods based on deep learning technologies can assist radiologists in achieving accurate reliable analysis size shape aneurysms, which may be helpful prediction models. However,...
Objective: To establish a CT image radiomics-based prediction model for the differential diagnosis of silicosis and tuberculosis nodules. Methods: A total 53 patients with 89 who underwent routine scans in Suzhou Fifth People's Hospital from January to August, 2018 were enrolled this study. AK/ITK software was used segment images obtain 139 lesions 119 lesions. For each lesion image, 396 features extracted, feature dimension reduction applied select most characteristic subset. Support vector...
Strategies to expand the pool of available donor organs include use extended criteria livers, which also injured allografts from donors with liver trauma.1 Trauma currently remains a leading cause death worldwide, and among those who sustain abdominal trauma, is most frequently organ. Hepatic injuries are seen in as many 40% patients sustaining trauma. Despite current trend toward conservative treatment, 14% all 33% major may require hepatic resection.2, 3 The deceased livers preexisting...
We study the problem of generating temporal object intrinsics -- temporally evolving sequences geometry, reflectance, and texture, such as a blooming rose from pre-trained 2D foundation models. Unlike conventional 3D modeling animation techniques that require extensive manual effort expertise, we introduce method generates assets with signals distilled diffusion To ensure consistency intrinsics, propose Neural Templates for temporal-state-guided distillation, derived automatically image...
Primary leiomyosarcoma of the inferior vena cava (IVC) is a rare malignant tumor originating from vein smooth muscle. We present one case primary IVC. The patient benefited surgical exploration at seventh day after admission. Tumor located in junction anterior wall IVC and left right renal vein. carried out resection, artificial vascular patch prosthetics. did not take anticoagulant drugs surgery was discharged 12 days surgery. Currently, had survived for nearly six months, repeated...
2D CT image-guided radiofrequency ablation (RFA) is an exciting minimally invasive treatment that can destroy liver tumors without removing them. However, images only provide limited static information, and the tumor will move with patient's respiratory movement. Therefore, how to accurately locate under free conditions urgent problem be solved at present.
The aim of this study was to establish an automatic classification model for chronic inflammation the choledoch wall using deep learning with CT images in patients pancreaticobiliary maljunction (PBM).
The effective classification of multi-task motor imagery electroencephalogram (EEG) is helpful to achieve accurate multi-dimensional human-computer interaction, and the high frequency domain specificity between subjects can improve accuracy robustness. Therefore, this paper proposed a EEG signal method based on adaptive time-frequency common spatial pattern (CSP) combined with convolutional neural network (CNN). characteristics subjects' personalized rhythm were extracted by spectrum...
The probabilistic classification vector machine (PCVM) is an effective sparse learning approach for binary classification. This paper presents extension of the PCVM to multiclass case, which aims provide reliable outputs by taking advantage inherent nature PCVM. performance proposed evaluated on several benchmark datasets. In particular, are used assess confidence in prediction. experimental results demonstrate that PCVMs can make more confident correct predictions than support machines and...
Computed tomography angiography (CTA) is a major noninvasive technology for imaging coronary artery disease, and effective accurate vessel tracking method can help radiologists diagnose the disease more accurately. In this paper, novel 3D vesse
BACKGROUND AND PURPOSE: Cerebral aneurysm is one of the most common cerebrovascular diseases, and SAH caused by its rupture has a very high mortality disability rate. Existing automatic segmentation methods based on DLMs with TOF-MRA modality could not segment edge voxels well, so that our goal to realize more accurate cerebral aneurysms in 3D help DLMs. MATERIALS METHODS: In this research, we proposed an framework TOF-MRA. The was composed two networks ranging from coarse fine. network,...