检索结果-南通市图书馆

Few-shot skin cancer detection based on many-objective optimization

在线全文

学校读者我要写书评

暂无评论

Kongzhi yu Juece/Control and Decision 2024年第11期39卷 3597-3606页

作者： Zhao, Jia-hui Wen, Jie Cai, Xing-juan Cui, Zhi-Hua Shanxi Key Laboratory of Big Data Analysis and Parallel Taiyuan University of Science and Technology Taiyuan 030024 China Shanxi Key Laboratory of Advanced Control and Equipment Ⅰntelligence Taiyuan University of Science and Technology Taiyuan 030024 China

Due to the long-tailed distribution characteristics of skin cancer data, quickly identifying rare skin disease samples with small amounts of data becomes a challenging few-shot problem. The detection method based on meta-learning can quickly learn meta-knowledge from the majority of common classes skin disease classes and use prior knowledge to improve the model’s ability to detect rare skin diseases. However, the bias in the quality and distribution of public skin cancer categories leads to the risk of overfitting in the pre-training phase of meta-learning, and meta-learning models based on traditional networks struggle to handle fine-grained skin disease problems. Affronting this issue, a many-objective meta-learning skin cancer detection model is proposed. The proposed model optimizes the distribution of common classes (base classes) by considering various classification performances of the skin cancer detection model, thus obtaining quality-enhanced training samples. And it adopts the ResNet12 network structure integrated with the CCNet attention model, which can significantly enhance the ability to identify fine-grained skin lesion images. Additionally,an improved many-objective optimization algorithm with discrete grouping strategy is designed to efficiently solve the proposed model. Some experiments are conducted on two public medical datasets, ⅠSCⅠ2018 and Derm7pt. The proposed many-objective skin detection model achieves detection accuracies of 67 %, 79 %, and 82 % respectively on binary classification tasks with 1-shot, 3-shot, and 5-shot learning,which validates the effectiveness of the model. © 2024 Northeast University. All rights reserved.

关键词： Contrastive Learning

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

基于多层特征表征与级联模型的医疗简历筛选

小型微型计算机系统 2022年第3期43卷 655-660页

作者：张晓莹陆慧娟中国计量大学现代科技学院杭州310018 中国计量大学信息工程学院杭州310018

本文研究医疗简历筛选,提出了基于多级特征表征与级联模型的方法.首先,面向医疗简历教育与工作背景、科研能力、业务能力等多组与多层特征表征问题,提出基于注意力网络的多级特征表征方法,实现特征级、特征组级与特征组组合等多级表征... 详细信息

本文研究医疗简历筛选,提出了基于多级特征表征与级联模型的方法.首先,面向医疗简历教育与工作背景、科研能力、业务能力等多组与多层特征表征问题,提出基于注意力网络的多级特征表征方法,实现特征级、特征组级与特征组组合等多级表征的融合.其次,提出了级联分类模型建模细粒度筛选过程,将现有简单的“通过与不通过”粗略筛选改进为“优秀、良好、中等、一般”的细粒度筛选,更加符合现今医院招聘初审、笔试、面试的多级筛选过程,有利于后续对不同等级人才的考量,也解决了数据不平衡带来的分类错误.最后,在包含13552条简历的数据集上验证了所提出方法,以支持向量机为分类方法的级联模型平均F1值达82.46%.特别是能够提升优秀类简历的分类效果,对可能被录用的竞聘者识别的更好,对于简历筛选的实际应用帮助更大.

关键词：医疗招聘简历筛选多级特征表征注意力网络级联分类模型

维普期刊数据库博看期刊

Predictive Ability of Hypertriglyceridemic Waist,Hypertriglyceridemic Waist-to-Height Ratio,and Waist-to-Hip Ratio for Cardiometabolic Risk Factors Clustering Screening among Chinese Children and Adolescents

在线全文

学校读者我要写书评

暂无评论

Biomedical and Environmental Sciences 2024年第3期37卷 233-241页

作者： XIAO Tian Li YUAN Shu Qian GAO Jing Yu Julien SBaker YANG Yi De WANG Xi Jie ZHENG Chan juan DONG Yan hui ZOU Zhi Yong Key Laboratory of Molecular Epidemiology of Hunan Province School of MedicineHunan Normal UniversityChangsha 410081HunanChina The Institute of Interdisciplinary Studies Hunan Normal UniversityChangsha 410013HunanChina Chenzhou First People's Hospital Chenzhou 423000HunanChina Centre for Health and Exercise Science Research Hong Kong Baptist UniversityKowloon TongHong Kong 999077China Department of Sport Physical Education and HealthHong Kong Baptist UniversityKowloon TongHong Kong 999077China Vanke School of Public Health Tsinghua UniversityBeijing 100084China Institute of Child and Adolescent Health School of Public HealthPeking University Health Science CenterBeijing 100191China

Objective Hypertriglyceridemic waist(HW),hypertriglyceridemic waist-to-height ratio(HWHtR),and waist-to-hip ratio(WHR)have been shown to be indicators of cardiometabolic risk ***,it is not clear which indicator is more suitable for children and *** aimed to investigate the relationship between HW,HWHtR,WHR,and cardiovascular risk factors clustering to determine the best screening tools for cardiometabolic risk in children and *** This was a national cross-sectional *** and biochemical variables were assessed in approximately 70,000 participants aged 6–18 years from seven provinces in ***,physical activity,dietary intake,and family history of chronic diseases were obtained through ***,χ2 and logistic regression analysis was *** A significant sex difference was observed for HWHtR and WHR,but not for HW *** risk of cardiometabolic health risk factor clustering with HW phenotype or the HWHtR phenotype was significantly higher than that with the non-HW or non-HWHtR phenotypes among children and adolescents(HW:OR=12.22,95%CI:9.54-15.67;HWHtR:OR=9.70,95%CI:6.93-13.58).Compared with the HW and HWHtR phenotypes,the association between risk of cardiometabolic health risk factors(CHRF)clustering and high WHR was much weaker and not significant(WHR:OR=1.14,95%CI:0.97-1.34).Conclusion Compared with HWHtR and WHR,the HW phenotype is a more convenient indicator with higher applicability to screen children and adolescents for cardiovascular risk factors.

关键词： Hypertriglyceridemic waist Waist-to-hip ratio Children and adolescents China Hypertriglyceridemic waist-to-height ratio Cardiovascular risk factors

Development and Validation of a Carbohydrate Metabolism-Related Model for Predicting Prognosis and Immune Landscape in Hepatocellular Carcinoma Patients

在线全文

学校读者我要写书评

暂无评论

Current Medical Science 2024年第4期44卷 771-788页

作者： Hong-xiang HUANG Pei-yuan ZHONG Ping LI Su-juan PENG Xin-jing DING Xiang-lian CAI Jin-hong CHEN Xie Zhui Zhi-hui luI Xing-yu TAO Yang-yang LIU Li CHEN Department of Oncology The First Affiliated Hospital of Nanchang UniversityNanchang330006China Jiangxi Institute of Respiratory Disease The First Affiliated Hospital of Nanchang UniversityNanchang330006China Department of Oncology Jiangxi Provincial People’s HospitalThe First Affiliated Hospital of Nanchang Medical CollegeNanchang330006China Department of Respiratory and Critical Care Medicine The First Affiliated Hospital of Nanchang UniversityNanchang330006China

Objective The activities and products of carbohydrate metabolism are involved in key processes of ***,its relationship with hepatocellular carcinoma(HCC)is *** The cancer genome atlas(TCGA)-HCC and ICGC-LIRI-JP datasets were acquired via public *** expressed genes(DEGs)between HCC and control samples in the TCGA-HCC dataset were identified and overlapped with 355 carbohydrate metabolism-related genes(CRGs)to obtain differentially expressed CRGs(DE-CRGs).Then,univariate Cox and least absolute shrinkage and selection operator(LASSO)analyses were applied to identify risk model genes,and HCC samples were divided into high/low-risk groups according to the median risk ***,gene set enrichment analysis(GSEA)was performed on the risk model *** sensitivity of the risk model to immunotherapy and chemotherapy was also *** A total of 8 risk model genes,namely,G6PD,PFKFB4,ACAT1,ALDH2,ACYP1,OGDHL,ACADS,and TKTL1,were ***,the risk score,cancer status,age,and pathologic T stage were strongly associated with the prognosis of HCC *** the stromal score and immune score had significant negative/positive correlations with the risk score,reflecting the important role of the risk model in immunotherapy ***,the stromal and immune scores had significant negative/positive correlations with risk scores,reflecting the important role of the risk model in immunotherapy ***,we found that high-/low-risk patients were more sensitive to 102 drugs,suggesting that the risk model exhibited sensitivity to chemotherapy *** results of the experiments in HCC tissue samples validated the expression of the risk model *** Through bioinformatic analysis,we constructed a carbohydrate metabolism-related risk model for HCC,contributing to the prognosis prediction and treatment of HCC patients.

关键词： hepatocellular carcinoma carbohydrate metabolism the cancer genome atlas bioinformatics

Different sperm sources and parameters can influence intracytoplasmic sperm injection outcomes before embryo implantation

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Zhejiang University-Science B(Biomedicine & Biotechnology) 2012年第1期13卷 1-10页

作者： Yue-hong lu1,2, hui-juan GAO1, Bai-jia LI1, Ying-ming ZHENG1, Ying-hui YE1, Yu-li QIAN1, Chen-ming XU1, He-feng HUANG1, Fan JIN1 (1Department of Reproductive Endocrinology, Key Laboratory of Reproductive Genetics, Ministry of Education, and Key Laboratory of Women’s Reproductive Health of Zhejiang Province, Women’s Hospital, School of Medicine, Zhejiang University, Hangzhou 310006, China) (2Department of Reproductive Health, Shaoxing Women and Children Hospital, Shaoxing 312000, China) Department of Reproductive Endocrinology Key Laboratory of Reproductive Genetics Ministry of Education and Key Laboratory of Women's Reproductive Health of Zhejiang Province Women's Hospital School of Medicine Zhejiang University Hangzhou 310006 China

To evaluate the effects of sperm with different parameters and sources on the outcomes of intracytoplasmic sperm injection (ICSI), 1972 ICSI cycles were analyzed retrospectively. Groups 1 to 5 were composed of cycles using ejaculated sperm and were grouped according to sperm quantity, quality, and morphology into normal (288 cycles), or mild (329 cycles), moderate (522 cycles), severe (332 cycles), and extremely severe (171 cycles) oligozoospermia and/or asthenozoospermia and/or teratozoospermia (OAT) groups. Group 6 was composed of 250 cycles using testicular or epididymal sperm, and Group 7 consisted of 80 cycles using frozen-thawed sperm. We found that fertilization rates were gradually reduced from Groups 1 to 6, and reached statistical difference in Groups 5 and 6 (P0.05). A total of nine cases of malformation were observed, with a malformation rate of 1.25% (9/719). In conclusion, different sperm sources and parameters can affect ICSI outcomes before embryo implantation. A full assessment of offspring mal-formation will require further study using a larger sample size.

关键词： Intracytoplasmic sperm injection (ICSI)i Sperm Sperm source Sperm parameters Malformation

在线全文

学校读者我要写书评

暂无评论

物流动态车辆调度问题的混合禁忌搜索算法

计算机工程与应用 2010年第8期46卷 228-231页

作者：王训斌陆慧娟张火明中国计量学院信息工程学院计算机系杭州310018

在分析动态车辆调度问题的基础上,建立了基于时间轴的动态模型;接着针对该问题在实际中的应用,设计了基于并行节约法和禁忌搜索的混合算法以对动态车辆调度问题进行求解;最后给出算法实现和算例模拟,验证了该算法的有效性。

关键词：带时窗动态车辆调度问题并行节约法禁忌搜索算法

Curcumin for gastric cancer:Mechanism prediction via network pharmacology,docking,and in vitro experiments

在线全文

学校读者我要写书评

暂无评论

World Journal of Gastrointestinal Oncology 2024年第8期16卷 3635-3650页

作者： Peng-hui Yang Ya-Nan Wei Bi-juan Xiao Si-Yi Li Xin-Long Li Liang-Jun Yang Hua-Feng Pan Geng-Xin Chen The Second School of Clinical Medicine Guangzhou University of Chinese MedicineGuangzhou 510006Guangdong ProvinceChina Science and Technology Innovation Center Guangzhou University of Chinese MedicineGuangzhou 510006Guangdong ProvinceChina Department of Traditional Chinese Medicine The People's Hospital of LonghuaShenzhen 518109Guangdong ProvinceChina Department of Gastroenterology Tongde Hospital of Zhejiang ProvinceHangzhou 310012Zhejiang ProvinceChina Centre for Translational Medicine The Second Affiliated Hospital of Guangzhou University of Chinese MedicineGuangzhou 510120Guangdong ProvinceChina

BACKGROUND Curcumin originates from the natural herb turmeric,and its antitumor effects have been known about for a long ***,the mechanism by which curcumin affects gastric cancer(GC)has not been *** To elucidate the potential mechanisms of curcumin in the treatment of *** Network pharmacological approaches were used to perform network analysis of *** first analyzed Lipinski’s Rule of Five for the use of *** latent targets were predicted using the PharmMapper,SwissTargetPrediction and DrugBank network *** disease targets were mined through the GeneCard,OMIM,DrugBank and TTD network ***,GO enrichment,KEGG enrichment,protein-protein interaction(PPI),and overall survival analyses were *** results were further verified through molecular docking,differential expression analysis and cell *** We identified a total of 48 curcumin-related genes with 31 overlapping GC-related *** intersection targets between curcumin and GC have been enriched in 81 GO biological processes and 22 significant *** PPI analysis,6 hub targets were identified,namely,estrogen receptor 1(ESR1),epidermal growth factor receptor(EGFR),cytochrome P450 family 3 subfamily A member 4(CYP3A4),mitogen-activated protein kinase 14(MAPK-14),cytochrome P450 family 1 subfamily A member 2(CYP1A2),and cytochrome p450 family 2 subfamily B member 6(CYP2B6).These factors are correlated with decreased survival rates among patients diagnosed with *** docking analysis further substantiated the strong binding interactions between Curcumin and the hub target *** experimental findings demonstrated that curcumin not only effectively inhibits the growth of BGC-823 cells but also suppresses their *** levels of hub targets CYP3A4,MAPK14,CYP1A2,and CYP2B6 in BGC-823 cells were significantly increased in each dose *** Curcumin can play an anti-GC role through a variety of

关键词： Curcumin Gastric cancer Network pharmacology Molecular docking Survival analysis

Multisystemic recurrent Langerhans cell histiocytosis misdiagnosed with chronic inflammation at the first diagnosis:A case report

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

World Journal of Radiology 2024年第6期16卷 232-240页

作者： Zhan-Rong Zhang Feng Chen hui-juan Chen Department of Radiology The Affiliated Hainan Hospital of Hainan Medical University(Hainan General Hospital)Haikou 570311Hainan ProvinceChina

BACKGROUND Langerhans cell histiocytosis(LCH)is characterized by diabetes insipidus and is an uncommon *** biopsies still have a certain degree of diagnostic *** present a case in which LCH initially affected the pituitary *** resulted in a misdiagnosis of chronic inflammation upon pathological *** SUMMARY A 25-year-old female exhibited symptoms of diabetes *** resonance imaging revealed an enhanced foci in the pituitary *** surgical resection of the pituitary lesion,the pathological diagnosis was chronic ***,the patient later experienced bone destruction in the skull and lower limb *** the lower limb bone lesion was compared with the initial pituitary lesion,the final diagnosis was modified to *** patient was treated with multiple chemotherapy ***,the patient’s condition gradually worsened,and she eventually passed away at *** LCH should be considered when patients exhibit diabetes insipidus and absence of high signal intensity in the pituitary gland on sagittal T1-weighted image and abnormal enhancement in the pituitary region.

关键词： Langerhans cell histiocytosis Pituitary gland Diabetes insipidus Multi-system Misdiagnosis Magnetic resonance imaging Case report

口腔冲洗用于气管插管危重患者口腔护理效果的meta分析

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

中华护理杂志 2013年第11期48卷 965-970页

作者：唐慧婷卢惠娟杨晓莉曹艳佩复旦大学护理学院上海市200032 复旦大学附属华山医院护理部

目的评价口腔冲洗应用于气管插管危重患者口腔护理的效果。方法计算机检索Cochrane Library、Joanna Briggs Institute Library、MEDLINE、RNAO、OVID、Web of Knowledge、Nursing Consult、EMbase、CINAHL、中国高等教育文献保障系统(CALIS)、中国生物医学文献服务系统(CBM)、中国期刊全文数据库(CNKI),以及万方数据资源系统中关于口腔冲洗应用于气管插管危重患者的随机对照试验、类实验性研究及队列研究,同时筛检纳入文献的参考文献。由2名研究者对文献质量进行严格评价和资料提取,对符合质量标准的研究进行meta分析。结果共纳入12项RCT及1项类实验性研究,1759例患者。meta分析结果证实,口腔冲洗应用于气管插管危重患者有助于降低呼吸机相关性肺炎(VAP)发生率[RR=0.46,95%CI(0.37,0.56),P<0.001],但单纯行口腔冲洗法降低VAP发生率的效果并不显著[RR=0.55,95%CI(0.21,1.47),P=0.24];冲洗与擦洗相结合的口腔护理方式能显著减少口腔并发症、VAP及口臭的发生[RR=0.16,95%CI(0.09,0.27),P<0.001;RR=0.40,95%CI(0.28,0.58),P<0.001;RR=0.09,95%CI(0.04,0.21),P<0.001];冲洗与刷洗结合亦有助于减少口腔并发症及VAP的发生[RR=0.34,95%CI(0.14,0.83),P=0.02;RR=0.49,95%CI(0.27,0.87),P=0.01]。结论对于气管插管的危重症患者而言,口腔冲洗联合擦洗或刷洗法可显著降低VAP及口腔并发症的发生率,改善ICU患者的口腔状况,建议临床推广使用。

关键词：插管法气管内口腔护理 Meta分析