A5069308334- Face and Expression Recognition
- Sparse and Compressive Sensing Techniques
- Remote Sensing and Land Use
- Domain Adaptation and Few-Shot Learning
- Bioinformatics and Genomic Networks
- Traditional Chinese Medicine Studies
- Spectroscopy and Chemometric Analyses
- Machine Learning and ELM
- Hormonal Regulation and Hypertension
- Spam and Phishing Detection
- Advanced Neural Network Applications
- Machine Learning in Bioinformatics
- Human Pose and Action Recognition
- Misinformation and Its Impacts
- Topic Modeling
- Traditional Chinese Medicine Analysis
- Tensor decomposition and applications
- Advanced Radiotherapy Techniques
- Text and Document Classification Technologies
- Spectroscopy Techniques in Biomedical and Chemical Research
- Hearing, Cochlea, Tinnitus, Genetics
- Machine Learning and Data Classification
- Remote-Sensing Image Classification
- Lung Cancer Treatments and Mutations
- Radiomics and Machine Learning in Medical Imaging
Shanghai University of Engineering Science
2025
Army Medical University
2022-2025
The Affiliated Yongchuan Hospital of Chongqing Medical University
2022-2025
Chongqing Medical University
2022-2025
Hong Kong Baptist University
2018-2024
Guangdong University of Technology
2024
Shaanxi University of Chinese Medicine
2023
State Key Laboratory of Biotherapy
2023
Sichuan University
2022-2023
Beijing Institute of Technology
2022
Automated annotation of protein function is challenging. As the number sequenced genomes rapidly grows, overwhelming majority products can only be annotated computationally. If computational predictions are to relied upon, it crucial that accuracy these methods high. Here we report results from first large-scale community-based critical assessment (CAFA) experiment. Fifty-four representing state art for prediction were evaluated on a target set 866 proteins 11 organisms. Two findings stand...
Sudden sensorineural hearing loss (SSHL) is a multifactorial disorder with high heterogeneity, thus the outcomes vary widely. This study aimed to develop predictive models based on four machine learning methods for SSHL, identifying best performer clinical application.Single-centre retrospective study.Chinese People's liberation army (PLA) hospital, Beijing, China.A total of 1220 in-patient SSHL patients were enrolled between June 2008 and December 2015.An advanced deep technique, belief...
Abstract Background Protein function determination is a key challenge in the post-genomic era. Experimental of protein functions accurate, but time-consuming and resource-intensive. A cost-effective alternative to use known information about sequence, structure, functional properties genes proteins predict using statistical methods. In this paper, we describe Multi-Source k -Nearest Neighbor (MS- NN) algorithm for prediction, which finds -nearest neighbors query based on different types...
Single-cell RNA sequencing has enabled to capture the gene activities at single-cell resolution, thus allowing reconstruction of cell-type-specific regulatory networks (GRNs). The available algorithms for reconstructing GRNs are commonly designed bulk RNA-seq data, and few them applicable analyze scRNA-seq data by dealing with dropout events cellular heterogeneity. In this paper, we represent joint expression distribution a pair as an image propose novel supervised deep neural network called...
Kernel support vector machines (SVMs) deliver state-of-the-art results in many real-world nonlinear classification problems, but the computational cost can be quite demanding order to maintain a large number of vectors. Linear SVM, on other hand, is highly scalable data only suited for linearly separable problems. In this paper, we propose novel approach called low-rank linearized SVM scale up kernel limited resources. Our transforms linear one via an approximate empirical map computed from...
Osteoporosis (OP) is a highly prevalent orthopedic condition in postmenopausal women and the elderly. Currently, OP treatments mainly include bisphosphonates, receptor activator of nuclear factor kappa-B ligand (RANKL) antibody therapy, selective estrogen modulators, teriparatide (PTH1-34), menopausal hormone therapy. However, increasing evidence has indicated these may exert serious side effects. In recent years, Traditional Chinese Medicine (TCM) become popular for treating disorders....
Non-small cell lung cancer (NSCLC) presents the most common type of cancer, accounting for 80-85% cases. Combining immunotherapy with radiotherapy (RT) has emerged as a significant research area in recent years. However, risk radiation pneumonitis, especially patients, poses concern. Iodine-125 (I125) seed implantation offers precise, less invasive alternative, minimizing damage to surrounding tissues and reducing side effects. This study aims evaluate safety efficacy I125 combined immune...
Abstract Intervention without implantation has become a requirement for developing percutaneous coronary intervention heart disease. In this paper, the recent advances of three representative types bioresorbable metal drug-eluting stents (DES) are reviewed, and material composition, structural design, mechanical properties degradability iron-based, magnesium-based zinc-based DES analyzed. The methods regulating radial strength degradation rate summarized, in vivo/in vitro performance...
When the amount of labeled data are limited, semisupervised learning can improve learner's performance by also using often easily available unlabeled data. In particular, a popular approach requires learned function to be smooth on underlying manifold. By approximating this manifold as weighted graph, such graph-based techniques achieve state-of-the-art performance. However, their high time and space complexities make them less attractive large sets. paper, we propose scale up set sparse...
The multiple instance regression (MIR) problem arises when a data set is collection of bags, where each bag contains instances sharing the identical real-valued label. goal to train model that can accurately predict label an unlabeled bag. Many remote sensing applications be studied within this setting. We propose novel probabilistic framework for MIR represents labels with mixture model. It based on assumption prime which responsible An expectation-maximization algorithm proposed maximize...
Auditory neuropathy spectrum disorder (ANSD) is one of the most common diseases leading to hearing and speech communication barriers in infants young children. The OTOF gene first identified for autosomal recessive non-syndromic ANSD, patients with mutations have shown marked improvement auditory functions from cochlear implantation, but true involvement Chinese ANSD still unknown which precludes effective management this disease. Here, we investigated contribution congenital China. In all,...
Leveraging information from the publicly accessible data repositories can be very useful when training a classifier small-sample microarray data. To achieve this, we proposed multi-task feature selection filter that borrows strength auxiliary It uses Kruskal?Wallis test on and ranks genes based their aggregated p-values. The top-ranked are selected as features for target task classifier. was evaluated related to nine different types of cancers. results showed is successful applied in...
Recent analyses have disclosed that existing rumor detection techniques, despite playing a pivotal role in countering the dissemination of misinformation on social media, are vulnerable to both white-box and surrogate-based black-box adversarial attacks. However, such attacks depend heavily unrealistic assumptions, e.g., modifiable user data access models, or appropriate selections surrogate which impractical real world. Thus, fail uncover robustness detectors practice. In this work, we take...
1. The objective of this study was to investigate the effects dietary probiotic supplementation on liver X receptor alpha (LXRα) and cholesterol 7α-hydroxylase (CYP7α1) mRNA levels, protein enzymatic activities fat metabolism in Cherry Valley Pekin ducks. 2. A total 750 one-day-old ducks were randomly divided into 5 groups with three replicates 50 each a completely randomised experiment. Each group fed basal diet supplemented 0, 500, 1000, 1500 or 2000 mg probiotics/kg. 3. Body rate feed...
Abstract Background Although the WHO Trial Registration Data Set (TRDS) has been published for many years, quality of clinical trial registrations with traditional Chinese medicine (TCM) is still not satisfactory, especially about inadequate reporting on TCM interventions. The development TRDS Extension 2020 (WHO TRDS-TCM 2020) aims to address this inadequacy. Methods A group experts, methodologists, epidemiologists, and editors developed through a comprehensive process, including baseline...
We demonstrate OceanRT, a novel cloud-based infrastructure that performs online analytics in real time, over large-scale temporal data such as call logs from telecommunication company. Apart proprietary systems for which few details have been revealed, most existing big-data are built on top of an offline, MapReduce-style infrastructure, inherently limits their efficiency. In contrast, OceanRT employs computing architecture consisting interconnected Access Query Engines (AQEs), well new...
Summary Diarrhoea caused by enterotoxigenic Escherichia coli (ETEC) expressing F4 (F4ab, F4ac and F4ad) fimbriae is a significant cause of mortality morbidity in newborn weaned pigs. The locus controlling susceptibility towards ETEC F4ab/ac has been mapped to SSC13q41, which TFRC (transferrin receptor) was localized considered as positional candidate gene for receptor. In this study, we determined susceptibility/resistance total 755 F 2 animals from White Duroc × Erhualian intercross using...
Convolutional neural networks (CNNs) have achieved significant performance on various real-life tasks. However, the large number of parameters in convolutional layers requires huge storage and computation resources, making it challenging to deploy CNNs memory-constrained embedded devices. In this article, we propose a novel compression method that generates convolution filters each layer using set learnable low-dimensional quantized filter bases. The proposed reconstructs by stacking linear...
Factorization Machines (FMs), a general predictor that can efficiently model high-order feature interactions, have been widely used for regression, classification and ranking problems. However, despite many successful applications of FMs, there are two main limitations FMs: (1) FMs consider interactions among input features by using only polynomial expansion which fail to capture complex nonlinear patterns in data. (2) Existing do not provide interpretable prediction users. In this paper, we...