A5100638725- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging and Analysis
- AI in cancer detection
- Visual perception and processing mechanisms
- Neural and Behavioral Psychology Studies
- Colorectal Cancer Screening and Detection
- Rough Sets and Fuzzy Logic
- Data Mining Algorithms and Applications
- Acute Ischemic Stroke Management
- Neural dynamics and brain function
- Dental Radiography and Imaging
- Advanced X-ray and CT Imaging
- Geoscience and Mining Technology
- Fuzzy Logic and Control Systems
- Energy Load and Power Forecasting
- Grey System Theory Applications
- Retinal Imaging and Analysis
- Image Retrieval and Classification Techniques
- Photoacoustic and Ultrasonic Imaging
- Thermoregulation and physiological responses
- Artificial Intelligence in Healthcare and Education
- Endometrial and Cervical Cancer Treatments
- Lung Cancer Diagnosis and Treatment
- Medical Imaging Techniques and Applications
- Advanced Power Amplifier Design
Hebei University
2015-2025
Innovation Team (China)
2024-2025
Chinese Academy of Medical Sciences & Peking Union Medical College
2023
Shandong University
2023
Shanghai Jiao Tong University
2016-2021
Colorectal cancer (CRC) represents one of the common malignancies gastrointestinal tract. The CRC incidence and mortality rates can be significantly reduced through early detection resection precursor lesions, also known colorectal polyps. However, such polyps missed during manual colonoscopy screening. With recent advances in artificial intelligence, numerous computer-aided diagnosis (CAD) methods have been proposed for applications. In particular, deep learning algorithms recently designed...
AI-based cervical lesion segmentation in colposcopy images has significant potential improving screening efficiency and accuracy. However, most current algorithms are insufficient for rapid image mass due to heavy parameters complex framework. Therefore, a lightweight algorithm real-time system is urgently needed. In this paper, novel LSIL + region framework termed Light-MDDNet proposed deployed, which uses the encoder-decoder architecture. encoder stage, first layer of MobileNetV2 module...
X-ray imaging is the primary diagnostic tool for clinical diagnosis of suspected fracture. Hand fracture (HF) a world-leading health problem children, adolescents and elderly. A missed hand on radiography may lead to severe consequences patients, resulting in delayed treatment poor recovery function. Nevertheless, many fractures are fairly insidious, which often misdiagnosis. In this dissertation, we propose GA_Faster R-CNN guided anchoring method (GA) GA_RPN applied detect localize...
Laser speckle contrast imaging (LSCI) is widely used for in vivo real-time detection and analysis of local blood flow microcirculation due to its non-invasive ability excellent spatial temporal resolution. However, vascular segmentation LSCI images still faces a lot difficulties numerous specific noises caused by the complexity microcirculation's structure irregular aberrations diseased regions. In addition, image data annotation have hindered application deep learning methods based on...
Objective: To investigate the magnetic resonance imaging (MRI) radiomics models in evaluating human epidermal growth factor receptor 2(HER2) expression breast cancer. Materials and Methods: The MRI data of 161 patients with invasive ductal carcinoma (non-special type) cancer were retrospectively collected, established based on features fat suppression T2 weighted image (T2WI) sequence, dynamic contrast-enhanced (DCE)-T1WIsequence joint sequences. T-test least absolute shrinkage selection...
Selective spatial attention enhances task performance at restricted regions within the visual field. The magnitude of this effect depends on level attentional load, which determines efficiency distractor rejection. Mechanisms load include perceptual selection and/or cognitive control involving working memory. Recent studies have provided evidence that microsaccades are influenced by attention. Therefore, microsaccade activities may be exploited to help understand dynamic selective under...
Abstract Colorectal cancer (CRC) is one of the main alimentary tract system malignancies affecting people worldwide. Adenomatous polyps are precursors CRC, and therefore, preventing development these lesions may also prevent subsequent malignancy. However, adenoma detection rate (ADR), a measure ability colonoscopist to identify remove precancerous colorectal polyps, varies significantly among endoscopists. Here, we attempt use convolutional neural network (CNN) generate unique...
The diagnosis of COVID-19 requires chest computed tomography (CT). High-resolution CT images can provide more diagnostic information to help doctors better diagnose the disease, so it is clinical importance study super-resolution (SR) algorithms applied improve resolution images. However, most existing SR are studied based on natural images, which not suitable for medical images; and these reconstruction quality by increasing network depth, machines with limited resources. To alleviate...
Objectives: To differentiate the primary small-cell lung cancer (SCLC) and non-small-cell (NSCLC) for patients with brain metastases (BMs) based on a deep learning (DL) model using contrast-enhanced magnetic resonance imaging (MRI) T1 weighted (T1CE) images. Methods: Out of 711 BMs origin (SCLC 232, NSCLC 479), MRI datasets 192 (lesions’ widths heights > 30 pixels) from (73 SCLC 119 NSCLC) confirmed pathologically were enrolled, retrospectively. A typical convolutional neural network...
To validate a radiomics model based on multi-sequence magnetic resonance imaging (MRI) in predicting the ki-67 expression levels early-stage endometrial cancer, 131 patients with early cancer who had undergone pathological examination and preoperative MRI scan were retrospectively enrolled divided into two groups levels. The features extracted from T2 weighted (T2WI), dynamic contrast enhanced T1 (DCE-T1WI), apparent diffusion coefficient (ADC) map screened using Pearson correlation...
Osteoporosis is a common systemic bone disease with insidious onset and low treatment efficiency. Once it occurs, will increase fragility lead to fractures. Computed tomography (CT) non-invasive medical examination method that can identify the condition of patients. In this paper, we propose novel channel attention module, which subsequently integrated into supervised deep convolutional neural network (DCNN) termed DSNet, perform feature fusion from two different scales, use quadratic weight...
Abstract Background The prevalence of hypertensive heart disease (HHD) is high and there currently no easy way to detect early HHD. Explore the application radiomics using cardiac magnetic resonance (CMR) non-enhanced cine sequences in diagnosing HHD latent changes caused by hypertension. Methods 132 patients who underwent CMR scanning were divided into groups: (42), hypertension with normal structure function (HWN) group (46), control (NOR) (44). Myocardial regions end-diastolic (ED)...
Endometrial carcinoma (EC) risk stratification prior to surgery is crucial for clinical treatment. In this study, we intend evaluate the predictive value of radiomics models based on magnetic resonance imaging (MRI) and staging early-stage EC. The study included 155 patients who underwent MRI examinations were pathologically diagnosed with EC between January, 2020, September, 2022. Three-dimensional features extracted from segmented tumor images captured by scans (including T2WI, CE-T1WI...
Abstract Preferentially processing behaviorally relevant information is vital for primate survival. In visuospatial attention studies, manipulating the spatial extent of focus an important question. Although many studies have claimed to successfully adjust field size by either varying uncertainty about target location (spatial uncertainty) or adjusting cue orienting focus, no systematic assessed and compared effectiveness these methods. We used a multiple paradigm with 2.5° 7.5° rings...
Diabetic retinopathy (DR) will cause blindness if the detection and treatment are not carried out in early stages. To create an effective strategy, severity of disease must first be divided into referral-warranted diabetic (RWDR) non-referral (NRDR). However, there usually no sufficient fundus examinations due to lack professional service communities, particularly developing countries. In this study, we introduce UGAN_Resnet_CBAM (URNet; UGAN is a generative adversarial network that uses...
Abstract Objective To establish a machine learning-based radiomics model to differentiate between glioma and solitary brain metastasis from lung cancer its subtypes, thereby achieving accurate preoperative classification. Materials methods A retrospective analysis was conducted on MRI T1WI-enhanced images of 105 patients with 172 cancer, which were confirmed pathologically. The divided into the training group validation in an 8:2 ratio for image segmentation, extraction, filtering; multiple...
Laser Speckle Contrast Imaging (LSCI) is an optical technology that provides exceptional spatio-temporal resolution and cost-effective capability for measuring blood flow. Extracting vascular features from LSCI images crucial in amplifying the imaging benefits assisting researchers analyzing local flow microcirculation. However, vessel segmentation has always been a technical challenge due to low signal-to-noise ratio of irregular morphology lesion areas. Supervised learning methods heavily...