A5090964683- Spectroscopy Techniques in Biomedical and Chemical Research
- Digital Imaging for Blood Diseases
- AI in cancer detection
- Remote-Sensing Image Classification
- Advanced Neural Network Applications
- Optical Imaging and Spectroscopy Techniques
- Spectroscopy and Chemometric Analyses
- Radiomics and Machine Learning in Medical Imaging
- COVID-19 diagnosis using AI
- Advanced Image and Video Retrieval Techniques
- Generative Adversarial Networks and Image Synthesis
- Image Processing Techniques and Applications
- Domain Adaptation and Few-Shot Learning
- Traditional Chinese Medicine Studies
- Advanced Image Fusion Techniques
- Medical Image Segmentation Techniques
- Image Retrieval and Classification Techniques
- Ultrasound in Clinical Applications
- Advanced Image Processing Techniques
- Infrared Thermography in Medicine
- Cell Image Analysis Techniques
- Non-Invasive Vital Sign Monitoring
- Remote Sensing and Land Use
- Colorectal Cancer Treatments and Studies
- Video Surveillance and Tracking Methods
East China Normal University
2016-2025
Linyi University
2025
Renji Hospital
2018-2024
Shanghai Jiao Tong University
2006-2024
Sichuan University
2017-2024
West China Second University Hospital of Sichuan University
2023-2024
Shanghai Municipal Education Commission
2018-2024
National Development and Reform Commission
2023-2024
Taishan University
2021-2024
Hebei Provincial Center for Disease Control and Prevention
2024
Research significance: The extended version of this paper has been accepted by IEEE Internet Things journal (DOI: 10.1109/JIOT.2020.2991456), please cite the version. During epidemic prevention and control period, our study can be helpful in prognosis, diagnosis screening for patients infected with COVID-19 (the novel coronavirus) based on breathing characteristics. According to latest clinical research, respiratory pattern is different from patterns flu common cold. One significant symptom...
In semi-supervised medical image segmentation, there exist empirical mismatch problems between labeled and un-labeled data distribution. The knowledge learned from the may be largely discarded if treating unlabeled separately or in an inconsistent manner. We propose a straightforward method for alleviating problem-copy-pasting bidirectionally, simple Mean Teacher architecture. encourages to learn comprehensive common semantics both inward outward directions. More importantly, consistent...
Blur was naturally analyzed in the frequency domain, by estimating latent sharp image and blur kernel given a blurry image. Recent progress on deblurring always designs end-to-end architectures aims at learning difference between pairs from pixel-level, which inevitably overlooks importance of kernels. This paper reveals an intriguing phenomenon that simply applying ReLU operation domain followed inverse Fourier transform, i.e., selection, provides faithful information about pattern (e.g.,...
In medical image analysis, the long-range spatial features are often not accurately obtained by traditional convolutional neural networks. Hence, we propose a TransClaw U-Net network structure. The transformer part is added after three convolution operations to fuse shallow extracted for maximally encoding between patches. "Claw" in U-net means that add bottom upsampling retain deepest feature information detail segmentation. addition, modified three-channel global attention mechanism blend...
Unpaired image-to-image (I2I) translation often requires to maximize the mutual information between source and translated images across different domains, which is critical for generator keep content prevent it from unnecessary modifications. The self-supervised contrastive learning has already been successfully applied in I2I. By constraining features same location be closer than those ones, implicitly ensures result take source. However, previous work uses random locations impose...
Pre-trained vision-language models like CLIP have recently shown superior performances on various downstream tasks, including image classification and segmentation. However, in fine-grained re-identification (ReID), the labels are indexes, lacking concrete text descriptions. Therefore, it remains to be determined how such could applied these tasks. This paper first finds out that simply fine-tuning visual model initialized by encoder CLIP, has already obtained competitive ReID Then we...
In patients who were treated with exogenous BMP-2 to repair bone fractures or defects, the levels of inflammatory cytokines such as TNF-α and IL-1β in sera are significantly elevated, which may affect outcome regeneration. Mitogen-activated protein kinase (MAPK) cascades extracellular signal-regulated 1/2 (ERK1/2), p38, c-Jun NH2-terminal (JNK1/2) have a crucial role osteogenic differentiation activated by both TNF-α/IL-1β. However, previous studies suggested that effects TNF-α/IL-1β...
A hyperspectral imaging system to measure and analyze the reflectance spectra of human tongue with high spatial resolution is proposed for tumor detection. To achieve fast accurate performance detecting tumors, data were collected using spectral acousto-optic tunable filters a adapter, sparse representation was used analysis algorithm. Based on image database, recognition rate 96.5% achieved. The experimental results show that diagnosis, together spectroscopic classification method provide...
Skin biopsy histopathological analysis is one of the primary methods used for pathologists to assess presence and deterioration melanoma in clinical. A comprehensive reliable pathological result correctly segmented its interaction with benign tissues, therefore providing accurate therapy. In this study, we applied deep convolution network on hyperspectral pathology images perform segmentation melanoma. To make best use spectral properties three dimensional data, proposed a 3D fully...
To solve the problem of supervised convolutional neural network (CNN) models suffering from limited samples, a two-channel CNN is developed for medical hyperspectral images (MHSI) classification tasks. In proposed network, one channel end-to-end denoted as EtoE-Net, designed to realize unsupervised learning, obtaining representative and global fused features with fewer noises, by building pixel-by-pixel mapping between two source data, i.e., original MHSI data its principal component. On...
Cell classification, especially that of white blood cells, plays a very important role in the field diagnosis and control major diseases. Compared to traditional optical microscopic imaging, hyperspectral imagery, combined with both spatial spectral information, provides more wealthy information for recognizing cells. In this paper, novel cell classification framework, which combines modulated Gabor wavelet deep convolutional neural network (CNN) kernels, named as MGCNN, is proposed based on...
Coronavirus Disease 2019 (COVID-19) has spread all over the world since it broke out massively in December 2019, which caused a large loss to whole world. Both confirmed cases and death have reached relatively frightening number. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), cause of COVID-19, can be transmitted by small droplets. To curb its at source, wearing masks is convenient effective measure. In most cases, people use face high-frequent but short-time way. Aimed...
Existing GAN inversion methods fail to provide latent codes for reliable reconstruction and flexible editing simultaneously. This paper presents a transformer-based image model pretrained StyleGAN which is not only with less distortions, but also of high quality flexibility editing. The proposed employs CNN encoder multi-scale features as keys values. Meanwhile it regards the style code be determined different layers generator queries. It first initializes query tokens learnable parameters...
In order to improve the field emission properties of screen-printed carbon nanotube films, ZnO nanostructures were deposited on multiwall nanotubes (MWNTs) using vapor phase transport method. Based combined effect geometrical structure MWNT/ZnO, a low turn-on ∼0.7V∕μm at current density 0.1μA∕cm2, threshold ∼2.3V∕μm 1mA∕cm2, high enhancement factor 8.2×103, and homogeneous image with spot ∼105cm−2 obtained from nanomultipods grown MNWT film.
Interleukin-17 (IL-17) is a proinflammatory cytokine that plays important roles in inflammation, autoimmunity, and cancer. The purpose of this study was to determine if IL-17 indirectly regulates macrophage differentiation through up-regulation cyclooxygenase-2 (COX-2) expression the cancer cell lines.Human cervical HeLa, human lung A549, mouse prostate Myc-CaP/CR lines were treated with recombinant IL-17; Western blot analysis, enzyme-linked immunosorbent assay, quantitative real-time...
This paper presents a real-time calibration method for gyro sensors in consumer portable devices. The happens automatically without the need external equipment or user intervention. Multi-level constraints, including pseudo-observations, accelerometer and magnetometer measurements, quasi-static attitude updates, are used to make reliable accurate under natural motions. Walking tests with Samsung Galaxy S3 S4 smartphones showed that estimate promising results even challenging motion modes...
Subjects with germline BRCA1/2 mutations (gBRCAm) have an increased risk of developing ovarian cancer and enhanced sensitivity to platinum-containing agents PARP (poly[ADP-ribose] polymerase) inhibitors. BRCA in Asian patients are poorly understood compared other populations. We aimed investigate gBRCAm prevalence characteristics Chinese patients.We conducted the first nationwide multicenter study China. Eight hundred twenty-six unselected from 5 clinical centers were enrolled tested for...
Microscopic examination is one of the most common methods for acute lymphoblastic leukemia (ALL) diagnosis.Most traditional automized blood cell identification are based on RGB color or gray images captured by light microscopes.This paper presents an method combining both spectral and spatial features to identify lymphoblasts from lymphocytes in hyperspectral images.Normalization encoding applied feature extraction support vector machinerecursive elimination (SVM-RFE) algorithm presented...