A5029578429- Machine Learning and ELM
- Face and Expression Recognition
- Lung Cancer Treatments and Mutations
- Iron and Steelmaking Processes
- Advanced Algorithms and Applications
- Brain Tumor Detection and Classification
- Neural Networks and Applications
- Domain Adaptation and Few-Shot Learning
- Conducting polymers and applications
- Organic Electronics and Photovoltaics
- Crystallization and Solubility Studies
- X-ray Diffraction in Crystallography
- Lung Cancer Research Studies
- Cancer Immunotherapy and Biomarkers
- PI3K/AKT/mTOR signaling in cancer
- Advanced Computational Techniques and Applications
- Mineral Processing and Grinding
- Crystallography and molecular interactions
- Metallurgical Processes and Thermodynamics
- Cancer, Lipids, and Metabolism
- Geoscience and Mining Technology
- Perovskite Materials and Applications
- Advanced Numerical Analysis Techniques
- Cancer Genomics and Diagnostics
- Cancer-related molecular mechanisms research
Liaoning Normal University
2024-2025
Xuzhou Medical College
2025
South China University of Technology
2014-2024
Dalian Polytechnic University
2024
Jiangsu Maritime Institute
2023
Daping Hospital
2019-2022
Army Medical University
2019-2022
Southwest Hospital
2021
Shaanxi University of Science and Technology
2020
Hunan University of Technology
2003-2019
Broad Learning System (BLS) that aims to offer an alternative way of learning in deep structure is proposed this paper. Deep and suffer from a time-consuming training process because large number connecting parameters filters layers. Moreover, it encounters complete retraining if the not sufficient model system. The BLS established form flat network, where original inputs are transferred placed as "mapped features" feature nodes expanded wide sense "enhancement nodes." incremental algorithms...
After a very fast and efficient discriminative broad learning system (BLS) that takes advantage of flatted structure incremental has been developed, here, mathematical proof the universal approximation property BLS is provided. In addition, framework several variants with their modeling given. The variations include cascade, recurrent, broad-deep combination structures. From experimental results, its outperform exist algorithms on regression performance over function approximation, time...
The broad learning system (BLS) has been proved to be effective and efficient lately. In this article, several deep variants of BLS are reviewed, a new adaptive incremental structure, Stacked BLS, is proposed. proposed model novel stacking BLS. This invariant inherits the efficiency effectiveness that structure weights lower layers fixed when blocks added. algorithm computes not only connection between newly but also enhancement nodes within block. considered as increment "layers" "neurons"...
This paper introduces a Broad Learning System that gives new paradigm and learning system without the need of deep architecture. In structure learning, abundant connecting parameters in filters layers lead to time-consuming training process. system, which is established as flat network, maps original inputs mapped features feature nodes expanded wide sense enhancement nodes. Model construction algorithms are introduced here. Moreover, different approaches for given. The advantage can be...
The fuzzy broad learning system (FBLS) is a recently proposed neuro-fuzzy model that shares the similar structure of (BLS). It shows high accuracy in both classification and regression tasks inherits fast computational nature BLS. However, ensemble several subsystems an FBLS decreases possibility understanding since rules from different systems are difficult to combine together while keeping consistence. To balance complexity, synthetically simplified with better interpretability, named...
The broad learning system (BLS) has recently been applied in numerous fields. However, it is mainly a supervised and thus not suitable for specific practical applications with mixture of labeled unlabeled data. Despite manifold regularization-based semi-supervised BLS, its performance still requires improvement, because assumption always applicable. Therefore, this article proposes an incremental-self-training-guided BLS (ISTSS-BLS). Distinctive to traditional self-training, where all data...
The broad learning system (BLS) is an algorithm that facilitates feature representation and data classification. Although weights of BLS are obtained by analytical computation, which brings better generalization higher efficiency, suffers from two drawbacks: 1) the performance depends on number hidden nodes, requires manual tuning, 2) double random mappings bring about uncertainty, leads to poor resistance noise data, as well unpredictable effects performance. To address these issues, a...
The current literature lacks reports on the roles of proliferative cells in tumorigenesis and causal relationship between cervical cancer. This study aims to investigate role mechanism Single-cell transcriptomics cancer were utilized identify cells. Mendelian randomization (MR) meta-analysis employed Additional assays such as 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), flow cytometry, gene set enrichment analysis (GSEA), weighted co-expression network (WGCNA)...
Broad Learning System [1] proposed recently demonstrates efficient and effective learning capability. This model is also proved to be suitable for incremental algorithms by taking the advantages of random vector flat neural networks. In this paper, a modified BLS structure based on K-means feature extraction developed. Compared with original broad system, acceptable performance more complicated data set, such as CIFAR-10, achieved. Furthermore, it that in flexible potential various applications.
Broad Learning System proposed recently [1] demonstrates efficient and effective learning capability. Moreover, fast incremental algorithms are developed in broad expansions without an entire retraining of the whole model. Compared with systems deep structure, inspired system provides competitive results classification. In this paper, applied to commonly used neural networks, such as radial basis function networks (RBF) hierarchical extremal machine (H-ELM). For RBF, resulting models, called...
Abstract Osimertinib, a 3rd generation epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI), is the first-line standard-of-care for EGFR-mutant non-small cell lung cancer (NSCLC) patients, while acquired drug resistance will inevitably occur. Interleukin-6 (IL-6) keystone cytokine in inflammation and cancer, its role osimertinib efficacy was unknown. Here we show that clinically, plasma IL-6 level predicts EGFR mutant NSCLC patients. Highly increased levels are found...
In recent years, research on facial expression recognition (FER) has become an increasingly active topic. Deep learning is a new area, which gives way to classify images of human faces into emotion categories. However, it many difficulties caused by poor robustness and real-time performance. This paper designs architecture network based Broad Learning System (BLS) for expressions recognition. It established as flat network. The original inputs are transferred placed mapped features in...
Abstract Background Immune‐therapy with anti‐PD1 inhibitors, such as pembrolizumab, is revolutionizing the treatment of non‐small cell lung cancers (NSCLC). However, identifying patients for potential therapeutic response and predicting therapy resistance early relapse remains a challenge. Methods Between 2016 2018, 60 were treated among who 12 NSCLC had both baseline (before treatment) serial (on periodical circulating tumor DNA (ctDNA) samples. Those samples sequenced on 329 pan...
It is hard to construct an optimal classifier for high-dimensional imbalanced data, on which the performance of classifiers seriously affected and becomes poor. Although many approaches, such as resampling, cost-sensitive, ensemble learning methods, have been proposed deal with skewed they are constrained by data noise redundancy. In this study, we propose adaptive subspace optimization method (ASOEM) classification overcome above limitations. To accurate diverse base classifiers, a novel...
Alzheimer's disease (AD) is a serious chronic health problem that causes great pain and loss to patients their families. Its early accurate diagnosis would achieve significant progress on the prevention treatment of disease. Magnetic Resonance Imaging (MRI) commonly used technique in nuclear medical diagnostics. However, it still challenging diagnose AD, Control Normal (CN), Mild Cognitive Impairment (MCI) because complex structures MRI. In this paper, diagnosing models for MRI images are...
Carbon‐containing pellets were prepared with the carbonized product of agricultural wastes and iron concentrate, an experimental study on direct reduction was carried out. The results demonstrated that carbon‐containing could be rapidly reduced at 1200 to 1300°C in 15 minutes, proper holding time high temperature 20 min. degree gradually increased rising, appropriate reducing 1200°C. weight loss rate rise carbon proportion, relatively reasonable mole ratio oxygen 0.9. A higher content...
Lung cancer ranks first both in morbidity and mortality malignancies, but prognostic biological markers are lacking. The neutrophil-lymphocyte ratio (NLR) was proposed as a convenient marker. This study aimed to explore the value of NLR advanced non-small cell lung (NSCLC).This retrospective screened patients admitted from October 2007 2014. Patients had histopathologically confirmed, treatment-naïve, metastatic NSCLC, were prescribed platinum doublet chemotherapy. demographic data...
Semi-supervised learning provides a solution to reduce the dependency of machine on labeled data. As one efficient semi-supervised techniques, self-training (ST) has received increasing attention. Several advancements have emerged address challenges associated with noisy pseudo-labels. Previous works acknowledge importance unlabeled data but not delved into their utilization, nor they paid attention problem high time consumption caused by iterative learning. This paper proposes Incremental...