A5056461668- T-cell and B-cell Immunology
- Immunotherapy and Immune Responses
- Hematopoietic Stem Cell Transplantation
- Electrocatalysts for Energy Conversion
- Intracerebral and Subarachnoid Hemorrhage Research
- AI in cancer detection
- Immune Cell Function and Interaction
- Optical Imaging and Spectroscopy Techniques
- Medical Image Segmentation Techniques
- CO2 Reduction Techniques and Catalysts
- Radiomics and Machine Learning in Medical Imaging
- Photoacoustic and Ultrasonic Imaging
- Acute Ischemic Stroke Management
- vaccines and immunoinformatics approaches
- Infrared Thermography in Medicine
- Optical Coherence Tomography Applications
- Neurosurgical Procedures and Complications
- Advanced Thermoelectric Materials and Devices
- RNA and protein synthesis mechanisms
- Advanced battery technologies research
- Ultrasound Imaging and Elastography
- Advanced Neuroimaging Techniques and Applications
- Acute Myeloid Leukemia Research
- Advanced MRI Techniques and Applications
- Ionic liquids properties and applications
Tencent (China)
2021-2025
Anhui University
2022-2025
The First Affiliated Hospital, Sun Yat-sen University
2025
Sun Yat-sen University
2025
Nanjing University of Science and Technology
2024-2025
Peking University
2017-2024
Peking University First Hospital
2017-2024
Changzhou University
2023
Shenzhen University Health Science Center
2018-2020
First Affiliated Hospital of Zhengzhou University
2016
ABUS, or Automated breast ultrasound, is an innovative and promising method of screening for examination. Comparing to common B-mode 2D ABUS attains operator-independent image acquisition also provides 3D views the whole breast. Nonetheless, reviewing images particularly time-intensive errors by oversight might occur. For this study, we offer convolutional network, which used automated cancer detection, in order accelerate meanwhile obtain high detection sensitivity with low false positives...
Abstract Pre-trained large language models demonstrate potential in extracting information from DNA sequences, yet adapting to a variety of tasks and data modalities remains challenge. To address this, we propose DNAGPT, generalized pre-training model trained on over 200 billion base pairs all mammals. By enhancing the classic GPT with binary classification task (DNA sequence order), numerical regression (guanine-cytosine content prediction), comprehensive token language, DNAGPT can handle...
Abstract Accurate prediction of antibody-antigen complex structures holds significant potential for advancing biomedical research and the design therapeutic antibodies. Currently, structure protein monomers has achieved considerable success, promising progress been made in extending this achievement to complexes. However, despite these advancements, fast accurate remains a challenging unresolved issue. Existing end-to-end methods, which rely on homology templates, exhibit sub-optimal...
Electroreduction of carbon dioxide into readily collectable and high-value carbon-based fuels is greatly significant to overcome the energy environmental crises yet challenging in development robust highly efficient electrocatalysts. Herein, a bismuth (Bi) heterophase electrode with enriched amorphous/crystalline interfaces was fabricated via cathodically situ transformation Bi-based metal-phenolic complexes (Bi-tannic acid, Bi-TA). Compared amorphous or crystalline Bi catalyst, structure...
Heterophase-boundary-abundant bismuth nanosheets were fabricated via facile electrochemical reduction of Bi-based coordination polymers, exhibiting excellent performance for CO 2 reduction.
Curved cobalt single atom catalysts could realize highly efficient CO 2 electroreduction, exhibiting industrial-level current density and high faradaic efficiency in pH-universal electrolytes.
Purpose Breast cancer is the most common and leading cause of cancer‐related deaths for women all over world. Recently, automated breast ultrasound (ABUS) has become a new promising screening modality whole examination. However, reviewing volumetric ABUS time‐consuming lesions could be missed during Therefore, computer‐aided detection in volume extremely expected to help clinician screening. Methods We develop novel end‐to‐end 3D convolutional network volume, order accelerate meanwhile...
Abstract Alpha-beta T cell receptor ( αβ TCR) recognition of peptide-major histocompatibility complexes (pMHCs) is a cornerstone the adaptive immune system. Fast and accurate modeling TCR-pMHC structures crucial for understanding TCR pMHCs at molecular level, which essential development TCR-based therapeutics vaccines. Despite significant interest, this challenge remains unresolved due to diversity interactions limited structural data. Here, we present tFold-TCR, high-throughput, end-to-end...
Abstract Single atom iron‐nitrogen‐carbon (Fe‐N‐C) catalysts with a planar Fe─N 4 structure are widely investigated as potential alternatives to platinum‐based materials for oxygen reduction reaction (ORR), while they still suffer from the imperfect adsorption and activation of intermediates, limiting their efficiency. Herein, Fe single‐atom catalyst biomimetic square pyramidal N 1 ‐Fe‐N site supported by honeycomb‐like porous carbon (SA‐FeN 5 /HPC) is successfully prepared supramolecular...
Acute myeloid leukemia (AML) is a prevalent and potentially fatal hematologic malignancy with limited improvements in survival rates over the past few decades. ITGAM, encoding CD11b, significant integrin component leukocytes, involved various biological processes. However, its role AML prognosis immune cell infiltration remains poorly understood. This study investigated prognostic significance potential function of ITGAM using comprehensive bioinformatic analyses. expression was markedly...
Abstract The role of the hydrogen bond network (HBN) within Helmholtz plane (HP) in regulating evolution kinetics for catalyst development remains ambiguous owing to lack fundamental understanding. Herein, leveraging ab initio molecular dynamics simulations, it is discovered that introducing weak metal bonds Ru/RuO 2 remarkably reshapes HBN. Subsequently, nanosheets loaded with single Ga atoms (Ga SA ‐Ru/RuO ) are successfully synthesized using a one‐step annealing strategy. In situ...
Accurately predicting the antigen-binding specificity of adaptive immune receptors (AIRs), such as T-cell (TCRs) and B-cell (BCRs), is essential for discovering new therapies. However, diversity AIR chain sequences limits accuracy current prediction methods. This study introduces SC-AIR-BERT, a pre-trained model that learns comprehensive sequence representations paired chains to improve binding prediction. SC-AIR-BERT first 'language' through self-supervised pre-training on large cohort from...
Pre-trained large language models demonstrate potential in extracting information from DNA sequences, yet adapting to a variety of tasks and data modalities remains challenge. To address this, we propose DNAGPT, generalized pre-training model trained on over 200 billion base pairs all mammals. By enhancing the classic GPT with binary classification task (DNA sequence order), numerical regression (guanine-cytosine content prediction), comprehensive token language, DNAGPT can handle versatile...
Lymphocyte trafficking via chemokine receptors such as C-C receptor 5 (CCR5) and CXCR3 plays a critical role in the pathogenesis of acute graft-versus-host disease (aGVHD). Our previous studies showed that addition CCR5 or antagonists could only slightly alleviate development aGVHD. Given specificity T lymphocytes bearing CCR5, we investigated whether combined blockade further attenuate murine A mouse model aGVHD was established to assess efficacy and/or on The distribution calculated by...
Abstract The central dogma serves as a fundamental framework for understanding the flow and expression of genetic information within living organisms, facilitating connection diverse biological sequences across molecule types. In this study, we present CD-GPT (Central Dogma Generative Pretrained Transformer), generative foundation model comprising 1 billion parameters, aiming to capture intricate system-wide molecular interactions in systems. We introduce concept unified representational...
Purpose Volumetric medical image registration has important clinical significance. Traditional methods may be time‐consuming when processing large volumetric data due to their iterative optimizations. In contrast, existing deep learning‐based networks can obtain the quickly. However, most of them require independent rigid alignment before deformable registration; these two steps are often performed separately and cannot end‐to‐end. Methods We propose an end‐to‐end joint affine network for...
Deformable image registration is of essential important for clinical diagnosis, treatment planning, and surgical navigation. However, most existing solutions require separate rigid alignment before deformable registration, may not well handle the large deformation circumstances. We propose a novel edge-aware pyramidal network (referred as EPReg) unsupervised volumetric registration. Specifically, we to fully exploit useful complementary information from multi-level feature pyramids predict...
Volumetric medical image registration has important clinical significance. Traditional methods may be time-consuming when processing large volumetric data due to their iterative optimizations. In contrast, existing deep learning-based networks can obtain the quickly. However, most of them require independent rigid alignment before deformable registration; these two steps are often performed separately and cannot end-to-end. Moreover, ground-truth is difficult for supervised learning methods....
We attempt to generate a definition of delayed perihematomal edema expansion (DPE) and analyze its time course, risk factors, clinical outcomes. A multi-cohort data was derived from the Chinese Intracranial Hemorrhage Image Database (CICHID). non-contrast computed tomography (NCCT) -based deep learning model constructed for fully automated segmentation hematoma (PHE). Time course PHE evolution correlated initial volume volumetrically assessed. Predictive values DPE were calculated through...