A5103004043- Radiomics and Machine Learning in Medical Imaging
- Advanced X-ray and CT Imaging
- Gastric Cancer Management and Outcomes
- Lung Cancer Diagnosis and Treatment
- AI in cancer detection
- Gastrointestinal Tumor Research and Treatment
- Colorectal Cancer Screening and Detection
- Medical Image Segmentation Techniques
- Medical Imaging Techniques and Applications
- Medical Imaging and Analysis
- MRI in cancer diagnosis
- Renal and Vascular Pathologies
- Urinary and Genital Oncology Studies
- Systemic Sclerosis and Related Diseases
- Advanced Computing and Algorithms
- Pancreatic and Hepatic Oncology Research
- Cervical Cancer and HPV Research
- Artificial Intelligence in Healthcare and Education
- COVID-19 diagnosis using AI
- Infrastructure Maintenance and Monitoring
- Pelvic and Acetabular Injuries
- Urological Disorders and Treatments
- Flow Measurement and Analysis
- Lung Cancer Treatments and Mutations
- Non-Destructive Testing Techniques
Beihang University
2022-2025
Chinese Academy of Sciences
2019-2024
Institute of Automation
2024
First Affiliated Hospital of Wenzhou Medical University
2024
Hefei University of Technology
2024
Wenzhou Medical University
2024
Shandong Institute of Automation
2016-2023
Ministry of Industry and Information Technology
2022-2023
Shaoxing People's Hospital
2023
University of Chinese Academy of Sciences
2019-2022
The Coronavirus disease 2019 (COVID-19) is raging across the world. radiomics, which explores huge amounts of features from medical image for diagnosis, may help screen COVID-19. In this study, we aim to develop a radiomic signature COVID-19 CT images. We retrospectively collect 75 pneumonia patients Beijing Youan Hospital, including 46 with and 29 other types pneumonias. These are divided into training set (n = 50) test 25) at random. segment lung lesions images, extract 77 lesions. Then...
Abstract Federated learning (FL) has shown great potential in addressing data privacy issues medical image analysis. However, varying distributions across different sites can create challenges aggregating client models and achieving good global model performance. In this study, we propose a novel personalized contrastive representation FL framework, named PCRFed, which leverages to address the non-independent identically distributed (non-IID) challenge dynamically adjusts distance between...
Providing precise and comprehensive diagnostic information to clinicians is crucial for improving the treatment prognosis of nasopharyngeal carcinoma. Multi-modal foundation models, which can integrate data from various sources, have potential significantly enhance clinical assistance. However, several challenges remain: (1) lack large-scale visual-language datasets carcinoma; (2) inability existing pre-training fine-tuning methods capture hierarchical features required complex tasks; (3)...
Objective: To develop and evaluate a diffusion-weighted imaging (DWI)-based radiomic nomogram for lymph node metastasis (LNM) prediction in advanced gastric cancer (AGC) patients. Overall Study: This retrospective study was conducted with 146 consecutively included pathologically confirmed AGC patients from two centers. All underwent preoperative 3.0 T magnetic resonance (MRI) examination. The dataset allocated to training cohort (n = 71) an internal validation 47) one center along external...
Encouraging and astonishing developments have recently been achieved in image-based diagnostic technology. Modern medical care imaging technology are becoming increasingly inseparable. However, the current diagnosis pattern of signal to image knowledge inevitably leads information distortion noise introduction procedure reconstruction (from image). Artificial intelligence (AI) technologies that can mine from vast amounts data offer opportunities disrupt established workflows. In this...
The potential prognostic value of extranodal soft tissue metastasis (ESTM) has been confirmed by increasing studies about gastric cancer (GC). However, the gold standard ESTM is determined pathologic examination after surgery, and there are no preoperative methods for assessment yet.
This study aimed to comprehensively evaluate non-contrast computed tomography (CT)-based radiomics for predicting early outcomes in patients with severe atherosclerotic renal artery stenosis (ARAS) after percutaneous transluminal angioplasty (PTRA). A total of 52 were retrospectively recruited, and their clinical characteristics pretreatment CT images collected. During a median follow-up period 3.7 mo, 18 confirmed have benefited from the treatment, defined as 20% improvement baseline...
Prognostic assessment remains a critical challenge in medical research, often limited by the lack of well-labeled data. In this work, we introduce ContraSurv, weakly-supervised learning framework based on contrastive learning, designed to enhance prognostic predictions 3D images. ContraSurv utilizes both self-supervised information inherent unlabeled data and cues present censored data, refining its capacity extract representations. For purpose, establish Vision Transformer architecture...
Abstract Background This study aimed to develope and validate a radiomics nomogram by integrating the quantitative characteristics of No.3 lymph nodes (LNs) primary tumors better predict preoperative node metastasis (LNM) in T1-2 gastric cancer (GC) patients. Methods A total 159 GC patients who had undergone surgery with lymphadenectomy between March 2012 November 2017 were retrospectively collected divided into training cohort (n = 80) testing 79). Radiomic features extracted from both...
Gastric cancer is one of the most common malignancies, ranking fifth in incidence and third mortality worldwide.1Smyth E.C. Nilsson M. Grabsch H.I. et al.Gastric cancer.Lancet. 2020; 396: 635-648Summary Full Text PDF PubMed Scopus (686) Google Scholar The high rate mainly caused by delayed early diagnosis inappropriate choice treatment. Neoadjuvant chemotherapy (NACT) combined with surgery recommended as routine treatment options for locally advanced gastric (LAGC).2Xu Q. Sun Z. Li X....
Abstract Objective. In the realm of utilizing artificial intelligence (AI) for medical image analysis, paradigm ‘signal-image-knowledge’ has remained unchanged. However, process ‘signal to image’ inevitably introduces information distortion, ultimately leading irrecoverable biases in ‘image knowledge’ process. Our goal is skip reconstruction and build a diagnostic model directly from raw data (signal). Approach . This study focuses on computed tomography (CT) its (sinogram) as research...
Tumor localization and lymph node metastasis (LNM) diagnosis are two important tasks for gynecologist to make decisions in cervical cancer treatments. Aiming develop an accurate convenient system, we propose a multi-task residual cross-attention network named MRCNet tumor segmentation LNM prediction. Specifically, tackle task correlation with underlying related supervision information, capture multi-level features by multi-scale convolutional neural network, which equipped module concerning...
In this paper, we investigate the problem of predicting histopathological findings gastric cancer (GC) from preoperative CT image. Unlike most existing classification systems assess global imaging phenotype tissues directly, formulate as a generalized multi-instance learning (GMIL) task and design deep GMIL framework to address it. Specifically, proposed aims at training powerful convolutional neural network (CNN) which is able discriminate informative patches neighbor confusing yield...
<h3>Background:</h3> Systemic lupus erythematosus, characterized by a severe autoimmune disease with strong individual heterogeneity, is particularly important to study the risk factors that affect its prognosis in order determine patient outcome and treatment. Currently, numerous studies have proven EBV infection can onset of SLE. Nevertheless, due diverse clinical manifestations SLE complex status EBV, it difficult predict prognostic patients systemic erythematosus infection....
The construction of prognostic prediction models based on follow-up data is crucial for devising individualized treatment plans patients. However, the performance current supervised survival analysis methods constrained due to prevalence weakly censored samples during follow-up. To address this limitation, study introduces Prognostic Co-Training Regression (PCTR) algorithm, a semi-supervised model developed through co-training two KNN regressors. By integrating prior information data, PCTR...
Patient-targeted treatment of non-small cell lung carcinoma (NSCLC) has been well documented according to the histologic subtypes over past decade. In parallel, recent development quantitative image biomarkers recently highlighted as important diagnostic tools facilitate histological subtype classification. this study, we present a radiomics analysis that classifies adenocarcinoma (ADC) and squamous (SqCC). We extract 52-dimensional, CT-based features (7 statistical 45 texture features)...
Abstract Recently, image-based diagnostic technology has made encouraging and astonishing development. Modern medical care imaging are increasingly inseparable. However, the current diagnosis pattern of Signal-to-Image-to-Knowledge inevitably leads to information distortion noise introduction in procedure image reconstruction (Signal-to-Image). Artificial intelligence (AI) technologies that can mine knowledge from vast amounts data offer opportunities disrupt established workflows. In this...
Accurately predict the risk of disease progression and benefit tyrosine kinase inhibitors (TKIs) therapy for stage IV non-small cell lung cancer (NSCLC) patients with activing epidermal growth factor receptor (EGFR) mutations by current staging methods are challenge. We postulated that integrating a classifier consisted multiple computed tomography (CT) phenotypic features, other clinicopathological factors into single model could improve stratification prediction progression-free survival...
Objective evaluation of medical image segmentation is one the important steps for proving its validity and clinical applicability. Although there are many researches presenting methods on image, while with few studying their results, this paper presents a learning method combined measures to make it as close possible clinicians' judgment. This more quantitative precise diagnose. In our experiment, same data sets include 120 results lumen-intima boundary (LIB) media-adventitia (MAB) carotid...