Article
Engineering, Aerospace
Seyyed Reza Ghaffari-Razin, Amir Reza Moradi, Navid Hooshangi
Summary: A new method for spatio-temporal modeling of ionosphere total electron content (TEC) using least squares support vector machine (LS-SVM) is proposed. The method reduces computational complexity, improves convergence speed and accuracy, and shows better performance in seasonal error analysis.
ADVANCES IN SPACE RESEARCH
(2022)
Article
Computer Science, Artificial Intelligence
Carlos Ruiz, Carlos M. Alaiz, Jose R. Dorronsoro
Summary: Multi-Task Learning aims to leverage the joint learning of a problem from both an overall and task-specific perspective. The proposed convex formulation for MTL typically outperforms the alternative optimal convex combination and the use of common or task-specific models alone.
Article
Automation & Control Systems
Di Song, Qianyi Wu, Mohammed Kamruzzaman
Summary: Informative spectral bands were selected from NIR spectroscopy to determine the proximate compositions in meat. Calibration models developed using these bands showed better prediction accuracy compared to individual bands and full spectral range, especially when non-linear calibration was used. The results suggest the effectiveness of common informative bands for predicting multiple quality attributes in meat.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2023)
Article
Environmental Sciences
Yanhui Li, Jiangjun Yao, Pengcheng Nie, Xuping Feng, Jizan Liu
Summary: The study found that using the LS-SVM model combined with terahertz spectral data to predict the degree of microplastic pollution in soil has a good effect, but seems to only be useful for local regions.
Article
Chemistry, Multidisciplinary
Thi Thanh Nha Tran, Thi Dieu Thuan Tran, Thi Thu Thuy Bui
Summary: The study introduces a novel modeling routine that combines machine learning techniques and tuning algorithms to improve the performance of 3D-CoMSIA models. The developed routine successfully mitigates the problem of overfitting and demonstrates superior performance compared to linear models. It also leads to the discovery of promising antioxidant peptides.
Article
Public, Environmental & Occupational Health
Hongfang Qiu, Han Zhao, Qi Chen, Qiyin Wang, Rong Ou, Mengliang Ye
Summary: Understanding the trend and seasonal patterns of varicella in Chongqing is crucial for risk analysis and healthcare resource allocation. The hybrid model, combining SARIMA and LS-SVM models, showed better prediction accuracy for varicella incidence, which peaks in April to June and October to December. The seasonal trend of varicella in Chongqing can be effectively predicted using the hybrid model analysis.
EPIDEMIOLOGY AND INFECTION
(2021)
Article
Neurosciences
Sunil Kumar Prabhakar, Dong-Ok Won
Summary: This research focuses on finger movement classification in the field of biomedical engineering and pattern recognition. Four proposed techniques for classifying sEMG signals are presented, including dynamic graph construction, dimensionality reduction with LTSA and LLC, and hybrid models using EA, BBN, ELM, DE, HFCM, and EWT. The best classification accuracy (98.5%) was achieved using the LMD-fuzzy C-means clustering technique with a combined kernel LS-SVM model.
FRONTIERS IN NEUROSCIENCE
(2023)
Article
Agronomy
Yuanyuan Liu, Tongzhao Wang, Rong Su, Can Hu, Fei Chen, Junhu Cheng
Summary: Customers are increasingly focusing on the sensory and physicochemical properties of food, leading to the use of near infrared hyperspectral technology to evaluate the quality parameters of Korla fragrant pears. By combining IRIV and LS-SVM, models were constructed to accurately assess the a*, firmness, and SSC of the fruit. The study demonstrated that a combination of spectral preprocessing methods, MSC and S-G, was the most effective in evaluating the quality parameters of the pears.
Article
Computer Science, Artificial Intelligence
Zhongbo Sun, Xin Zhang, Keping Liu, Tian Shi, Jing Wang
Summary: This paper proposes an active motion intention recognition technology based on LS-SVM and ZNN, which successfully estimates the continuous motion angles of knee joint and hip joint. By processing the sEMG signal and establishing a noise-suppressing neural network, accurate identification and noise elimination are achieved, providing valuable reference for lower limb joint movement of patients.
NEURAL PROCESSING LETTERS
(2023)
Article
Plant Sciences
Huixue Dong, Suli Yan, Yexing Jing, Ruizhen Yang, Yunwei Zhang, Yun Zhou, Yingfang Zhu, Jiaqiang Sun
Summary: The miR156-targeted SPLs enhance ABA responses and inhibit sprouting, with SPL9 directly activating ABA responsive genes by interacting with ABI5 and SnRK2s.
FRONTIERS IN PLANT SCIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Anand Kumar Srivastava, Yugal Kumar, Pradeep Kumar Singh
Summary: In the medical field, researchers have proposed the K-Mean(++) data imputation technique and ABC-based outlier detection technique to address missing values and outliers in medical datasets. The performance of the proposed hybrid diabetes prediction framework using LS-SVM classifiers has been evaluated with high accuracy, sensitivity, specificity, kappa, and AUC rates, outperforming 34 state-of-the-art techniques.
Article
Environmental Sciences
Huu-Tuan Tran, Mohammed Hadi, Thi Thu Hang Nguyen, Hong Giang Hoang, Minh-Ky Nguyen, Khoi Nghia Nguyen, Dai-Viet N. Vo
Summary: This study investigates the potential of using easily measurable parameters to predict the quantities of microplastics in peatland sediment samples. The results demonstrate that the quantity of microplastics is primarily influenced by pH, TOC, and salinity. Additionally, three machine learning models achieved considerable accuracy when applied to the testing dataset.
MARINE POLLUTION BULLETIN
(2023)
Article
Chemistry, Analytical
Ahmed Youssef Ali Amer, Femke Wouters, Julie Vranken, Pauline Dreesen, Dianne de Korte-de Boer, Frank van Rosmalen, Bas C. T. van Bussel, Valerie Smit-Fun, Patrick Duflot, Julien Guiot, Iwan C. C. van der Horst, Dieter Mesotten, Pieter Vandervoort, Jean-Marie Aerts, Bart Vanrumste
Summary: This study introduces machine learning predictive models to predict the future values of the monitored vital signs of COVID-19 ICU patients. By limiting the number of vital sign predictors and increasing the observation rate, the prediction performance can be enhanced.
Article
Engineering, Marine
Tianqi Pei, Caoyang Yu, Yiming Zhong, Lian Lian, Xianbo Xiang
Summary: In this study, a novel approach for maneuvering prediction of marine crafts in data-limited situations is proposed. The approach combines least square-support vector machines (LS-SVM) with a self-error corrector that integrates K-means clustering with Markov model. The proposed strategy shows superiority over the conventional LS-SVM method in terms of prediction performance, reducing the root mean square error of surge speed by up to 86.2% in simulation tests and 54.5% in experiments.
Article
Chemistry, Applied
Rani Amsaraj, Sarma Mutturi
Summary: This study successfully demonstrated the simultaneous detection and quantification of four color adulterants in black tea using FT-IR spectral data and chemometric tools.
JOURNAL OF FOOD COMPOSITION AND ANALYSIS
(2024)
Article
Chemistry, Multidisciplinary
Lu Zhang, Xian-meng Li, Xu-he Shi, Kai Ye, Xue-li Fu, Xue Wang, Shi-man Guo, Jia-qi Ma, Fei-fei Xu, Hui-min Sun, Qian-qian Li, Wei-ying Zhang, Li-hong Ye
Summary: HBXIP prevents sorafenib-induced cell death in hepatocellular carcinoma and attenuates ferroptosis.
ACTA PHARMACOLOGICA SINICA
(2023)
Article
Rheumatology
Jun Li, Suli Wang, Jiayi Zheng, Qianqian Li, Jia Li, Liangjing Lu
Summary: This study investigated the risk of PJP in newly diagnosed MDA5 + DM patients within 6 months and found a high incidence and mortality rate in these patients, with certain treatment methods being associated with a higher risk.
CLINICAL RHEUMATOLOGY
(2023)
Letter
Rheumatology
Hanlin Yin, Bei Liu, Qianqian Li, Qingran Yan, Liangjing Lu
ARTHRITIS & RHEUMATOLOGY
(2023)
Article
Chemistry, Applied
Yu Huang, Chunyuan Liang, Yueling Cai, Yi Zhou, Bingkun Guo, Jipeng Cheng, Heguang Liu, Peng Wang, Qianqian Li, Anmin Nie, Hongtao Wang, Jinsong Wu, Tongyi Zhang
Summary: The fundamental reaction behaviors of nickel sulfide (NixSy) as lithium-ion battery anodes were investigated using in-situ TEM. Ni3S2 was found to be the electrochemically stable phase. During lithiation, NixSy nanoparticles evolved into a mixture of Ni nanocrystals embedded in Li2S matrix, forming a porous structure. In delithiation, the Ni3S2 phase recrystallized to form a compact and interconnected network. Carbon substrate improved the tolerance for size-dependent pulverization of NixSy nanoparticles. This work enhances the understanding of electrochemical reaction behavior and aids in the rational design of high-energy density battery anodes.
JOURNAL OF ENERGY CHEMISTRY
(2023)
Article
Chemistry, Multidisciplinary
Wenkang Miao, Ronghui Hao, Jingzhou Wang, Zihan Wang, Wenxin Lin, Heguang Liu, Zhenjie Feng, Yingchun Lyu, Qianqian Li, Dongling Jia, Runhai Ouyang, Jipeng Cheng, Anmin Nie, Jinsong Wu
Summary: Non-noble metal catalysts have been successfully designed for the hydrogenation of CO2, with high conversion activity and stability. The reversible regeneration mechanism of catalytic activity has been revealed. This study is of great significance for rational structure design and improving catalyst performance.
Review
Chemistry, Multidisciplinary
Yang Zhou, Qianqian Li, Ye Wu, Xinyu Li, Ya Zhou, Zhu Wang, Hui Liang, Feiqing Ding, Sheng Hong, Nicole F. Steinmetz, Hui Cai
Summary: Self-assembly is a widely used method for constructing biomaterials, and peptides have been extensively investigated due to their desirable properties. Stimuli-responsive peptide nanoparticles, capable of conformational and chemical changes in response to stimuli, have emerged as a promising class of materials with diverse biomedical applications.
Article
Chemistry, Physical
Ronghui Hao, Wenkang Miao, Qiling Xiao, Zihan Wang, Wanyin Xu, Tianzhong Wang, Qianqian Li, Peng Wang, Anmin Nie, Hongtao Wang
Summary: This study reports the oxidation pathways of TiN with different grain sizes through careful atomic structure characterization. Nano-sized TiN single crystal undergoes a two-step phase transformation, while micron-sized TiN single crystal or polycrystals directly convert to rutile TiO2 upon oxidation. N2 nanobubbles are generated in the micron-sized TiN oxidation process, which preferentially embed inside of rutile TiO2 grain boundaries for single crystals and the oxide/nitride interface for polycrystals. The oxidation induces great volume expansion in all types of TiN, but only micron-sized single crystals are fractured and pulverized dramatically, impairing the mechanical properties.
APPLIED SURFACE SCIENCE
(2023)
Article
Forestry
Wenyi Zhang, Yanrong Yang, Cheng Hu, Leying Zhang, Bo Hou, Weifeng Wang, Qianqian Li, Yansong Li
Summary: With climate change, forest fires are becoming more frequent worldwide, resulting in increased carbon emissions and affecting the carbon cycle of forest ecosystems. However, the impact of forest fires on carbon cycles varies by region, and there is a lack of comparative studies. This research examines the carbon loss under forest fire disturbances in northeast and southwest natural forest areas of China, using NPP data as an important parameter of forest carbon absorption. The study reveals differences in forest fire characteristics, carbon emissions, and the impact on forest NPP between the two regions.
Article
Pharmacology & Pharmacy
Yu-Ting Xiong, Jian-Fei Wang, Xiao-Xia Niu, Yi-Ming Fu, Ke-Xin Wang, Chun-Yan Wang, Qian-Qian Li, Jian-Jun Wang, Jun Zhao, Dong Ji
Summary: Clinical characteristics and risk factors associated with the severity of drug-induced liver injury (DILI) in elderly patients were identified. The presence of autoimmunity was found to be associated with a more serious state of DILI.
FRONTIERS IN PHARMACOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Qianqian Li, Wenyi He, Weiran Li, Shulu Luo, Minghong Zhou, Dingcai Wu, Yan Li, Shuyi Wu
Summary: In guided bone regeneration surgery, a new class of local double-layered adhesive barrier membranes (ABMs) has been developed to immobilize bone graft materials. These ABMs have suction-adhesion properties and can firmly adhere to wet bone surfaces, effectively preventing displacement and leakage. This versatile strategy may open up new avenues for designing next-generation barrier membranes.
Article
Immunology
Ruixin Song, Chao Yang, Qianqian Li, Jiayin Wang, Jing Chen, Kai Sun, Hongmin Lv, Yankai Yang, Jing Liang, Qing Ye, YanYing Gao, Jun Li, Ying Li, Junqing Yan, Ying Liu, Tao Wang, Changen Liu, Ping Zhu, Fei Wang, Weili Yin, Huiling Xiang
Summary: The aim of this study was to evaluate the durability of immune response after basic and booster immunization with SARS-CoV-2 vaccines in patients with chronic liver disease (CLD). The results showed that the positive rates of antibodies decreased rapidly with the extension of vaccination time after basic immunization, while the positive rates increased rapidly and remained high after booster immunization. Thus, booster immunization improved the immune response and significantly prolonged the durability of SARS-CoV-2 antibodies in patients with CLD.
FRONTIERS IN IMMUNOLOGY
(2023)
Article
Immunology
Qingran Yan, Mei Zhang, Fang Du, Yuening Kang, Ping Ye, Qianqian Li, Bei Liu, Min Dai, Chunde Bao
Summary: This prospective study evaluated the efficacy and safety of IGU as an add-on therapy for patients with refractory lupus nephritis. The results showed that IGU, as a component of combination therapy, can effectively reduce urine protein and is well tolerated.
FRONTIERS IN IMMUNOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Shuoshuo Li, Wenyu Hu, Shenghui Gong, Ping Zhang, Jinbo Cheng, Shukun Wang, Yingyi Wang, Wenjun Shi, Qianqian Li, Fengchao Wang, Zengqiang Yuan
Summary: High altitude exposure leads to cognitive impairments by affecting the cerebral vasculature system and reducing oxygen and nutrition supply to the brain. The study investigates the molecular mechanism of vascular system remodeling under acute hypoxia and identifies a novel m6A reader protein, PRRC2B, in endothelial cells. Knockdown of PRRC2B promotes hypoxia-induced endothelial cell migration and vascular remodeling, alleviating hypoxia-induced cognitive decline. These findings provide a new potential therapeutic target for hypoxia-induced cognitive impairments.
Article
Computer Science, Interdisciplinary Applications
Qianqian Li, Jingqi Zeng, Lijuan Ma, Jinyuan Zhu, Fangyu Zhang, Yunan Wei, Mingshuang Li, Lu Yao, Jing Du, Zhisheng Wu
Summary: In this study, an industrial information integrated strategy was proposed to address the OQC evaluation issue of Chinese meteria medica (CMM) from multi-batch in the complex system of big honeyed pill. The strategy consisted of a two-step digital process control and an OQC diagnosis strategy. The two-step digital control utilized online near infrared (NIR) monitoring to acquire critical quality attributes (CQAs) and control consistency from honey refining to He-Tuo procedure. The OQC diagnosis strategy utilized data fusion approach to determine multi-batch OQC from different perspectives. The information integrated strategy successfully identified three abnormal samples out of 30 batches with weights of 27.82%, 44.60%, and 27.58% for NIR, mid-infrared (MIR), and texture technologies, respectively. The diagnosed samples were also tested using zebrafish efficacy as an in vivo test, which demonstrated the effectiveness of the OQC diagnosis strategy. Overall, the proposed information integrated strategy provides important theoretical and practical foundations for ensuring the overall quality of multi-batch CMM, especially for digital and intelligent industrialization.
JOURNAL OF INDUSTRIAL INFORMATION INTEGRATION
(2023)
Article
Chemistry, Physical
Hailin Shen, Yueling Cai, Zhongtao Ma, Peng Wang, Bingkun Guo, Jipeng Cheng, Qianqian Li, Hongtao Wang, Zhongyuan Liu, Anmin Nie, Jinsong Wu
Summary: The researchers report a metal phosphorous trichalcogenide of MnPS3 as a reliable and high-capacity anode material for lithium-ion batteries. It has a unique layered van der Waals structure and can enable fast insertion/extraction of ions.