检索结果-南通市图书馆

Prediction of sepsis within 24 hours at the triage stage in emergency departments using machine learning

在线全文

学校读者我要写书评

暂无评论

World Journal of Emergency Medicine 2024年第5期15卷 379-385页

作者： Jingyuan Xie Jiandong Gao Mutian Yang Ting Zhang Yecheng liu Yutong Chen Zetong liu qimin Mei Zhimao Li Huadong Zhu Ji Wu Department of Electronic Engineering Tsinghua UniversityBeijing 100084China Center for Big Data and Clinical Research Institute for Precision MedicineTsinghua UniversityBeijing 100084China Institute of Medical Technology Peking University Health Science CenterBeijing 100091China Emergency Department State Key Laboratory of Complex Severe and Rare DiseasesPeking Union Medical College HospitalChinese Academy of Medical Science and Peking Union Medical CollegeBeijing 100005China Department of Health Care Peking Union Medical College HospitalChinese Academy of Medical Sciences and Peking Union Medical CollegeBeijing 100005China

BACKGROUND:Sepsis is one of the main causes of mortality in intensive care units(ICUs).Early prediction is critical for reducing *** approximately 36%of sepsis occur within 24 h after emergency department(ED)admission in Medical Information Mart for Intensive Care(MIMIC-IV),a prediction system for the ED triage stage would be *** methods such as the quick Sequential Organ Failure Assessment(qSOFA)are more suitable for screening than for prediction in the ED,and we aimed to fi nd a light-weight,convenient prediction method through machine ***:We accessed the MIMIC-IV for sepsis patient data in the *** dataset comprised demographic information,vital signs,and synthetic *** Gradient Boosting(XGBoost)was used to predict the risk of developing sepsis within 24 h after ED ***,SHapley Additive exPlanations(SHAP)was employed to provide a comprehensive interpretation of the model's *** percent of the patients were randomly selected as the testing set,while the remaining patients were used for training with 10-fold ***:For 10-fold cross-validation on 14,957 samples,we reached an accuracy of 84.1%±0.3%and an area under the receiver operating characteristic(ROC)curve of 0.92±*** model achieved similar performance on the testing set of 1,662 *** values showed that the fi ve most important features were acuity,arrival transportation,age,shock index,and respiratory ***:Machine learning models such as XGBoost may be used for sepsis prediction using only a small amount of data conveniently collected in the ED triage *** may help reduce workload in the ED and warn medical workers against the risk of sepsis in advance.

关键词： Sepsis Machine learning Emergency department Triage Informatics

Immune-tumor interaction dictates spatially directed evolution of esophageal squamous cell carcinoma

在线全文

学校读者我要写书评

暂无评论

National Science Review 2024年第5期11卷 170-185页

作者： Yong Zhou Shanlan Mo Heyang Cui Ruifang Sun Weimin Zhang Xiaofei Zhuang Enwei Xu Hongyi Li Yikun Cheng Yongsheng Meng Meilin liu Ting Yan Huijuan liu Ling Zhang Bin Yang Yanfeng Xi Shubin Wang Xiaolong Cheng Shuai Cheng Li Zhihua liu qimin Zhan Zheng Hu Yongping Cui Cancer Institute Department of Pathology Peking University Shenzhen Hospital Shenzhen Peking University-the Hong Kong University of Science and Technology(PKU-HKUST) Medical Center Institute of Cancer ResearchShenzhen Bay Laboratory Key Laboratory of Cellular Physiology of the Ministry of Education Department of Pathology Shanxi Medical University City University of Hong Kong Shenzhen Research Institute CAS Key Laboratory of Quantitative Engineering Biology Shenzhen Institute of Synthetic BiologyShenzhen Institute of Advanced TechnologyChinese Academy of Sciences Department of Surgery Faculty of Medicine The Chinese University of Hong Kong Department of Tumor Biobank Shanxi Province Cancer HospitalShanxi Hospital Affiliated to Cancer HospitalChinese Academy of Medical SciencesCancer Hospital Affiliated to Shanxi Medical University Key Laboratory of Carcinogenesis and Translational Research(Ministry of Education/Beijing) Laboratory of Molecular Oncology Peking University Cancer Hospital & Institute Research Unit of Molecular Cancer Research Chinese Academy of Medical Sciences Department of Thoracic Surgery Shanxi Province Cancer HospitalShanxi Hospital Affiliated to Cancer HospitalChinese Academy of Medical SciencesCancer Hospital Affiliated to Shanxi Medical University Department of Pathology Shanxi Province Cancer HospitalShanxi Hospital Affiliated to Cancer HospitalChinese Academy of Medical SciencesCancer Hospital Affiliated to Shanxi Medical University College of Letters & Science University of California Berkeley State Key Laboratory of Molecular Oncology National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital Chinese Academy of Medical Sciences and Peking Union Medical College

Esophageal squamous cell carcinoma（ESCC） is a poor-prognostic cancer type with extensive intra-and inter-patient heterogeneity in both genomic variations and tumor microenvironment（TME）. However, the patterns and drivers of spatial genomic and microenvironmental heterogeneity of ESCC remain largely unknown. Here, we generated a spatial multi-omic atlas by whole-exome, transcriptome, and methylome sequencing of 507 tumor samples from 103 patients. We identified a novel tumor suppressor PREX2,accounting for 22% of ESCCs with frequent somatic mutations or hyper-methylation, which promoted migration and invasion of ESCC cells in vitro. Analysis of the TME and quantification of subclonal expansion indicated that ESCCs undergo spatially directed evolution, where subclones mostly originated from the tumor center but had a biased clonal expansion to the upper direction of the ***, we found upper regions of ESCCs often underwent stronger immunoediting with increased selective fitness, suggesting more stringent immune selection. In addition, distinct TMEs were associated with variable genomic and clinical outcomes. Among them, hot TME was associated with high immune evasion and subclonal heterogeneity. We also found that immunoediting, instead of CD8+T cell abundance, acts as an independent prognostic factor of ESCCs. Importantly, we found significant heterogeneity in previously considered potential therapeutic targets, as well as BRCAness characteristics in a subset of patients, emphasizing the importance of focusing on heterogeneity in ESCC targeted ***, these findings provide novel insights into the mechanisms of the spatial evolution of ESCC and inform precision therapeutic strategies.

关键词： squamous cell carcinoma heterogeneity multi-omics evolution tumor microenvironment

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

邯钢5号高炉铜冷却壁生产实践

炼铁 2008年第2期27卷 5-7页

作者：赵贵希刘琪敏邯郸钢铁集团公司

结合邯钢5号高炉炉墙结厚的处理,总结了铜冷却壁高炉维护合理操作炉型的经验。认为加强入炉原燃料的管理,减少成分波动,降低入炉粉末含量,通过上下部调剂维持适当的边缘煤气流,以及选择合适的热制度、造渣制度是炉况稳定顺行的关键。

关键词：大型高炉铜冷却壁结厚操作制度

图像跨尺度特征融合与数据增强的公交客流检测模型

在线全文

学校读者我要写书评

暂无评论

武汉大学学报（信息科学版） 2024年第5期49卷 700-708页

作者：黄丽龚志鹏刘凡凡程起敏中南民族大学电子信息工程学院湖北武汉430074 华中科技大学电子信息与通信学院湖北武汉430074

针对目前公交客流目标检测中由于前后排乘客图像尺寸差距大、遮挡严重而难以检测的问题,提出一种改进的目标检测模型YOLOv5s_P。该模型基于YOLOv5基本架构,针对乘客图像尺寸差距大的问题,使用重复加权双向特征金字塔网络结构替换YOLOv5... 详细信息

针对目前公交客流目标检测中由于前后排乘客图像尺寸差距大、遮挡严重而难以检测的问题,提出一种改进的目标检测模型YOLOv5s_P。该模型基于YOLOv5基本架构,针对乘客图像尺寸差距大的问题,使用重复加权双向特征金字塔网络结构替换YOLOv5中的路径聚合网络结构,引入权重机制双向跨尺度融合不同特征,增强提取复杂目标特征的能力;同时针对遮挡问题,结合Mixup数据增强方法,扩充遮挡和重叠图像的训练样本,提高模型泛化能力,减少客流图像残缺导致的漏检。为了验证YOLOv5s_P模型性能,将其应用于实际公交场景,并与Faster R-CNN、SSD300、RetinaNet、YOLOv5s 4种模型进行对比测试。实验结果表明,YOLOv5s_P模型在不影响检测速度的情况下,平均精度均值达到96.9%,平均漏检率较YOLOv5s模型降低了3.43%,提高了公交客流的检测精度。

关键词：深度学习 YOLOv5s_P 公交客流目标检测特征融合数据增强

俄制500MW汽轮机中压缸膨胀及胀差问题研究

在线全文

学校读者我要写书评

暂无评论

华北电力技术 2000年第9期 22-24,43页

作者：邱爱玲刘启民华北电力设计院北京100011 天津盘山发电厂天津301900

汽轮机在启动、暖机和升速、停机过程中 ,或在运行中工况发生变化 ,或机组负荷发生变化时 ,汽轮机缸体与转子间的热膨胀值均发生变化。通过对俄制К- 50 0 - 2 4 0 - 4型 50 0 MW机组结构的介绍 ,根据监测装置的安装情况及机组整体试验... 详细信息

汽轮机在启动、暖机和升速、停机过程中 ,或在运行中工况发生变化 ,或机组负荷发生变化时 ,汽轮机缸体与转子间的热膨胀值均发生变化。通过对俄制К- 50 0 - 2 4 0 - 4型 50 0 MW机组结构的介绍 ,根据监测装置的安装情况及机组整体试验和运行情况 ,结合机组缸内动、静部分设计间隙值 ,经过分析、试验、计算及实际运行情况的观察 ,得出结论 ,可以将俄制 К- 50 0 - 2 4 0 - 4型汽轮机组中压缸胀差值由设计值 +4.5～ - 2 .0 mm扩展为 +5.5～ - 2 .0 mm,机组内部动静部分不会磨擦 ,可以保证机组安全长期运行。

关键词：膨胀胀差中压缸汽轮机 К-500-240-4型

在线全文

学校读者我要写书评

暂无评论

600MW汽轮机高压内缸螺栓可靠性研究

能源与节能 2015年第4期 168-171,178页

作者：刘启民神华国华(北京)电力研究院北京100025

通过对亚临界机组高压内缸的受力情况分析,对内缸密封螺栓强度与应力进行有限元计算,借鉴同型机组高压内缸螺栓故障原因与螺栓材料性能分析,根据螺栓材料实际运行工况下持续强度与应力及高温蠕变的曲线趋势,分析了螺栓应力变化对内缸密... 详细信息

通过对亚临界机组高压内缸的受力情况分析,对内缸密封螺栓强度与应力进行有限元计算,借鉴同型机组高压内缸螺栓故障原因与螺栓材料性能分析,根据螺栓材料实际运行工况下持续强度与应力及高温蠕变的曲线趋势,分析了螺栓应力变化对内缸密封安全可靠性的影响,并对机组检修间隔优化提供了参考依据。

关键词：高压内缸载荷应力持久强度高温蠕变

在线全文

学校读者我要写书评

暂无评论

火力发电厂锅炉给水泵振动问题分析与研究

能源与节能 2015年第6期 134-137页

作者：刘启民神华国华(北京)电力研究院北京100069

从火力发电厂运行角度出发,根据机组负荷及系统介质参数的变化,分析了锅炉给水泵在正常运行过程中发生振动的原因及影响最佳运行参数与最大效率的工况点,结合锅炉给水泵的结构特点、泵体内部损失及转子动力的变化,分析了泵体振动的原因... 详细信息

从火力发电厂运行角度出发,根据机组负荷及系统介质参数的变化,分析了锅炉给水泵在正常运行过程中发生振动的原因及影响最佳运行参数与最大效率的工况点,结合锅炉给水泵的结构特点、泵体内部损失及转子动力的变化,分析了泵体振动的原因及防范措施,使得锅炉给水泵的可用率有效提高。

关键词：振动寿命效率汽蚀推力

在线全文

学校读者我要写书评

暂无评论

论基层医院护理文件质量控制

护理学杂志 2005年第6期20卷 58-59页

作者：詹静刘奇敏南阳市中心医院护理部河南南阳473009 南阳市中心医院生殖中心

分析基层医院在护理文件质量控制上存在方法落后、监督体系不健全、职责不清、奖惩不力、赏罚不明等问题。提出必须建立护理文件的逐级负责制;将责、权、利统一起来;改进质控方法,建立健全质量控制网络体系;组建一支高质量的护理管理队... 详细信息

分析基层医院在护理文件质量控制上存在方法落后、监督体系不健全、职责不清、奖惩不力、赏罚不明等问题。提出必须建立护理文件的逐级负责制;将责、权、利统一起来;改进质控方法,建立健全质量控制网络体系;组建一支高质量的护理管理队伍;不断提高护理人员的业务能力。

关键词：基层医院护理文件质量控制

描摹时代光影,或写给世界的书信——文珍访谈录

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

写作 2023年第5期43卷 5-10页

作者：刘启民文珍湖南省社会科学院文学所老舍文学院

80后作家文珍一直对当代城市生活抱有敏锐而持久的关注,城市里各异的人甚至“非人”都成为文珍观照的对象,体他人之心式的写作背后展现的强大共情力,让文珍成为青年写作者中独特的存在。在采访中,文珍回顾了自己的阅读史和写作史,谈到... 详细信息

80后作家文珍一直对当代城市生活抱有敏锐而持久的关注,城市里各异的人甚至“非人”都成为文珍观照的对象,体他人之心式的写作背后展现的强大共情力,让文珍成为青年写作者中独特的存在。在采访中,文珍回顾了自己的阅读史和写作史,谈到了曾经做田野调查的经验与写作体会,同时提到一个作家不同文体创作之间的联系与区别。文珍意识到一位力图书写时代的创作者进入社会深处的重要性,但对她来说,不同于社会学家总是在外部观视,作家最好更切身地成为“不动声色的隐形观察者”,去真正感受现场的气息与温度。

关键词：文珍阅读史城市书写文体