Analysis on the results of graduate examination for standardized medical residency training in general hospital---- Takes The First Affiliated Hospital of Kunming Medical University as an example

[ 目的] 对昆明医科大学第一附属医院（ 以下简称昆医附一院）2016 -2017年住院医师规范化培训结 业综合考核结果进行分析和总结,查找培训过程中存在的问题,提出提高住院医师规范化培训质量的建议. [方法] 将云南省住院医师规范化培训管理平台中公布的昆医附-院2016 - 2017年住院医师规范化培训结业综 合考核的结果录入Excel, 按照考试站点各专业考核情况进行分类统计,并对数据进行分析.[结果] 各站点考 试的合格率分别为综合知识部分95.26%,辅助检查部分88.23%,病史采集部分97.64%,体格检查部分 96. 14% ,病例分析部分97.21%,临床操作部分94.21%,门诊病例书写部分97.42%,医患沟通部分98.07%. 各专业综合考核第-阶段的不合格率：最高的为泌尿外科63. 64 %,最低的为内科学6.25%.各专业综合考核 第二阶段的不合格率：最高的为神经内科33.33%,最低的为超声科6. 67 %.[结论] 结业考核结果中,除辅助 检查部分外,其余七站均达到90 %以上的合格率,说明辅助检查判读的部分是该院参培住院医师的薄弱环节, 应加强检查结果判读培训,以确保提高住院医师规范化培训的综合质量.

Research on the Application of the OSCE Model in Vocational Nursing Comprehensive Practical Training

The objective of this study is to examine the application effect of the Objective Structured Clinical Examination(OSCE)model in the comprehensive practical training course for higher vocational nursing.Based on the evaluation dimensions of core nursing competencies,the OSCE framework was constructed with the clinical nursing work process as the main focus and supported by a nursing case database.A questionnaire survey revealed that over 80%of students in the experimental group were satisfied with the application of OSCE in the nursing comprehensive practical training course,and more than 90%believed that OSCE assessment enhanced their theoretical knowledge and practical skills.The OSCE was also considered helpful in transforming learned knowledge and skills into professional competencies.

The Role of the Examination Management for Medical Students in Humanities Education and Training Professional Quality

It has been proved by years of research that if the function of examination management cannot be fully understood or be used effectively, the humanities education and train of professional quality for medical students will not make a fundamental achievement. Because it is not only an academic training but also the examination management and the humanities cultivation have a close connection. The humanities cultivation which is based on the core of virtue requires a good campus environment. Therefore, we should adjust to the medical students＇ humanities quality training requirements, making the targeted reform and innovation of examination management.

Application of Serum CK and BUN Determination in Monitoring Pre-Competition Training of Badminton Athletes

In order to investigate the feasibility of serum creatine kinase （CK） and blood urea nitrogen （BUN） in monitoring pre-competition training of badminton athletes, the pre-competition training load of 20 badminton athletes was studied, and serum CK and BUN were determined before, immediate and next morning after training. The results showed that after intensive training for one week, serum CK levels were significantly increased by 57.53 mmol/L （P〈0.05）. After regulation of the training intensity, average serum CK levels were increased by 21.79 mmol/L （P〈0.05）. BUN contents were increased by 0.83 mmol/L on average with the difference being not significant （P〉0.05）. After intermittent training, there was significant difference in the average increased levels of serum CK in athletes （P〈0.05）. There was significant difference before and after regulation of training （P〈0.05）. The increased levels of BUN were 0.78 mmol/L without significant difference （P〉0.05）. It was concluded that serum CK was one of the biochemical indicators monitoring the training load sensitivity of badminton athletes, but BUN was of little value in monitoring the training load. Both serum CK and BUN recovered slowly after one-week intensive training and intermittent training, suggesting the metabolic mechanism of human body in training needs further study.

Effective distributed convolutional neural network architecture for remote sensing images target classification with a pre-training approach

How to recognize targets with similar appearances from remote sensing images(RSIs) effectively and efficiently has become a big challenge. Recently, convolutional neural network(CNN) is preferred in the target classification due to the powerful feature representation ability and better performance. However,the training and testing of CNN mainly rely on single machine.Single machine has its natural limitation and bottleneck in processing RSIs due to limited hardware resources and huge time consuming. Besides, overfitting is a challenge for the CNN model due to the unbalance between RSIs data and the model structure.When a model is complex or the training data is relatively small,overfitting occurs and leads to a poor predictive performance. To address these problems, a distributed CNN architecture for RSIs target classification is proposed, which dramatically increases the training speed of CNN and system scalability. It improves the storage ability and processing efficiency of RSIs. Furthermore,Bayesian regularization approach is utilized in order to initialize the weights of the CNN extractor, which increases the robustness and flexibility of the CNN model. It helps prevent the overfitting and avoid the local optima caused by limited RSI training images or the inappropriate CNN structure. In addition, considering the efficiency of the Na¨?ve Bayes classifier, a distributed Na¨?ve Bayes classifier is designed to reduce the training cost. Compared with other algorithms, the proposed system and method perform the best and increase the recognition accuracy. The results show that the distributed system framework and the proposed algorithms are suitable for RSIs target classification tasks.

Pre/Post Assessments Analysis in Training Electro-Technical Seafarers Experts

In maritime industry, personnel’s training is considered by shipping companies as a top priority matter on the list of factors affecting competitiveness in operating vessels. This paper presents the importance and the effects of training Electro-Technical Experts in the context of latest developments, particularly the advent of the “Electric Ship” and the “Communicative Ship” analyzing the feedback received from several relevant two-days seminars for “Ship Electrical and Electronic Systems for Electro-Technical Officers”, in North East European countries. The pre-test and post-test self assessment method that has been used for more efficient interaction between trainers and trainees is analyzed using t-statistics. The attendees have had diverse basic backgrounds, yet company experts Fleet Engineers on merchant or war ships. The training’s effectiveness and gain is discussed in this paper and further proposals for the Electrical and Electronic training are presented through the valuable feedback for improvement.

Predictors of Performance of the Maritime Academy of Asia and the Pacific （MAAP） in the OIC Navigational Watch Licensure Examination

Abstract： The study aimed to assess the predictors of performance administered by the Philippine Regulation Commission （PRC） in the licensure examination of Maritime Academy of Asia and the Pacific （MAAP） for Academic Year 2007-2011. Specifically, this also aimed to answer the following questions： demographic profile of the maritime graduates, the college related factors, administrative related factors, the level of scholastic achievement of the graduates, and the performance of the graduates in the licensure examination. A total of one hundred seventy three （173） graduates of MAAP from Class 2007-2011 were utilized as the respondents for the study. The study found out that the results of pre board examination and scholastic achievement of the graduates had significant effect on the performance of the graduates in O.I.C Navigational Watch Licensure Examination. Other variables found out to have no significant effect on the result of the licensure examination of the graduates. Therefore, the null hypotheses that the demographic profile of the graduates, college related factors, and administrative related factors has no significant effect on scholastic the achievement of the students were accepted. However, the null hypothesis that scholastic achievement of the students has no significant effect on the performance in maritime licensure examination was rejected. The same way, the null hypothesis that the demographic profile of the graduates, college related factors and the administrative related factors has no significant effect on the performance in Officer in Charge--Navigational Watch licensure Examination were accepted.

Outcomes of the Expanded Programme on Immunization Pre-Service Training Initiatives in Kenya: A Mixed Methods Study

Background: The Maternal and Child Survival Program of United States Agency for International Development conducted a study in 2017 to assess the outcome of an initiative to strengthen Expanded Programme on Immunization (EPI) pre-service training. The pre-service training initiative was undertaken by the Ministry of Health (MOH) with support from partners in 2012-2016. The overall objective of the study was to assess the adoption and effectiveness of the initiative in the competency (knowledge, skills and attitude) of graduate nurses. Methods: The study included a conveniently selected sample of 14 pre-service training institutions, 23 field practicum sites, and 29 health facilities in western Kenya, and used quantitative and qualitative methods of data collection. Results: All pre-service training institutions were found to have adapted the WHO EPI prototype curriculum. Overall, tutors followed training method in the classroom as suggested in the curriculum, except evaluation of students’ learning lacked tests or quizzes. Students had opportunities for hands-on practical experience in the field practicum sites. Graduate nurses were found to have acquired the skills for vaccinating children. However, some pre-service training institutions lacked functional skills labs for practical learning of students. In addition, students did not receive up-to-date information on EPI program, and lacked knowledge and skills on monitoring and documentation of EPI coverage during preservice training. Conclusions: It appears that the EPI pre-service training strengthening initiatives facilitated competency-based EPI training of nurses in Kenya. However, preservice training institutions still have scope for improvement in the skills lab, hand-washing practice, providing up-to-date information, and training students on coverage monitoring and documentation.

Analysis of the Effect of Systematic Pre-Job Training for Nurses in Isolation Wards during the COVID-19 Pandemic

Objective: To explore the effect of systematic pre-job training for isolation ward nurses during the Corona Virus Disease 2019 (COVID-19) pandemic. Methods: Establish a pre-job training program for the isolation ward for COVID-19, standardize the content of theoretical and skill training, formulate training SOPs, and conduct training for the nurses using online teaching assessment, video teaching, on-site scenario simulation operation drills, as well as real-time protection guidance and supervision. 60 nurses from non-infectious departments temporarily selected by the hospital were trained;the theoretical knowledge scores, quarantine techniques, and nursing quality of nurses before and after the training were compared, and the effect of the intervention was evaluated. Results: The scores of the COVID-19 protection theory test were 81.17 ± 8.46 after the nurses were trained for 3 days. The pass rates of hand hygiene compliance tests and protective clothing putting-on and taking-off practices were 96.67% and 100%, respectively. There was no significant difference between the scores of the COVID-19 protection theory test for the nurses that were trained for 3 days and the scores for the nurses originally at the quarantine zone (81.59 ± 7.59, P > 0.05). The pass rate of hand hygiene compliance and the pass rate of protective clothing putting-on and taking-off practices were significantly improved compared with those before training (81.67% and 56.67% respectively, P < 0.001). The scores of the COVID-19 protection theory test at 30 days of training were 95.67 ± 5.89, which were significantly higher than those at 3 days of training (P < 0.001). The qualified rate of disinfection and quarantine in the first month for the trained nurses entering the isolation ward was 89.47%;compared with that for the nurses originally in the isolation wards (94.7%), there was no significant difference (P > 0.05). The comprehensive nursing ability scores of bedside nurses in the first month of training were 80.14 ± 5.63, which were lower than those of nurses originally in the isolation wards (86.88 ± 4.53, P Conclusion: Systematic pre-job training for nurses in isolation wards can help improve nurses’ knowledge of the COVID-19, self-protection awareness, and protection skills, and can quickly train nurses who are competent for work in isolation wards. It is an important guarantee of “zero infection” for medical staff, and it can quickly and effectively help medical institutions respond to the COVID-19 pandemic in an emergency.

Prefix-LSDPM:面向小样本的在线学习会话退出预测模型

在线学习会话退出预测旨在准确预测在线学习过程中的学习会话退出,是智慧教育领域中十分重要的一项研究任务。针对现有模型在小样本场景下预测准确率较低的问题,提出了基于前缀提示的在线学习会话退出预测模型Prefix-LSDPM。该模型为获取单个学习行为内部特征及连续学习行为之间的隐含关联信息,在改进了键值向量的Transformer网络中对提示形式的合成序列进行掩码学习;为降低模型训练涉及的参数量以适应小样本学习,将学习会话退出预测任务建模形式靠近预训练任务,并在冻结的预训练参数基础上对提示参数进行调优。基于多个数据集的实验结果表明,Prefix-LSDPM的预测准确率优于现有模型,且在小样本学习中仍能达到较好的预测效果。

自适应特征融合的多模态实体对齐研究

多模态数据间交互式任务的兴起对于综合利用不同模态的知识提出了更高的要求,因此融合不同模态知识的多模态知识图谱应运而生.然而,现有多模态知识图谱存在图谱知识不完整的问题,严重阻碍对信息的有效利用.缓解此问题的有效方法是通过实体对齐进行知识图谱补全.当前多模态实体对齐方法以固定权重融合多种模态信息,在融合过程中忽略不同模态信息贡献的差异性.为解决上述问题,设计一套自适应特征融合机制,根据不同模态数据质量动态融合实体结构信息和视觉信息.此外,考虑到视觉信息质量不高、知识图谱之间的结构差异也影响实体对齐的效果,本文分别设计提升视觉信息有效利用率的视觉特征处理模块以及缓和结构差异性的三元组筛选模块.在多模态实体对齐任务上的实验结果表明,提出的多模态实体对齐方法的性能优于当前最好的方法.

基于RoBERTa-Span-Attack的标签指针网络军事命名实体识别

军事领域文本中存在大量军事实体信息,准确识别这些信息是军事文本信息提取和构建军事知识图谱的基础性任务。首先,提出了一种基于RoBERTa预训练模型、跨度和对抗训练的标签指针网络的融合深度模型(RoBERTa-Span-Attack),用于中文军事命名实体识别;然后,采用了一种基于Span的标签指针网络,同时完成实体的起止位置和类别的识别任务;最后,在模型训练过程中加入对抗训练策略,通过添加一些扰动来生成对抗样本进行训练。在军事领域数据集上的实验结果表明:所提出的军事领域命名实体识别模型相较于BERT-CRF、BERT-Softmax和BERT-Span,在识别准确度上具有更优的效果。

“经师”与“人师”统一:我国中小学教师职前培养的关键难点

中小学教师职前培养是其职业生涯的主体中心环节,在此过程中强调“经师”与“人师”的统一,是其可否成为教师职业胜任者的重中之重,也是中小学教师职前培养的关键难点。当前我国中小学教师职前培养偏重“经师”,体现出知识本位、技能本位、实践本位等目标取向。实现“经师”与“人师”的统一培养特别是培养“人师”意义重大,培养“人师”不仅是弘扬中华优秀传统文化的内在要义,而且是落实“立德树人”教育根本任务的必然要求,还是新时代建设高质量教师队伍的根本遵循。目前在实现“经师”与“人师”统一过程中,还面临着培养主体育人理念存在错位、培养对象在专业学习过程中功利化倾向较为严重、培养过程的核心维度和关键内容失之偏颇等问题。基于此,中小学教师职前培养应遵循国家聚焦制度政策建设、学校致力培养过程改革、高校教师夯实“人师”培养能力、师范生主动提升“人师”力等多维协同,最终实现“经师”与“人师”的统一培养。

中医医院新入职本科护士培训考核评价指标的构建

目的 构建中医医院新入职本科护士培训考核评价指标。方法 通过文献研究、小组讨论,建立中医医院新入职本科护士培训考核评价指标初稿,选取来自7个省份的中华护理学会中医中西医结合护理分会的15名专家进行2轮专家咨询,确定考核评价内容及其权重。结果 2轮专家函询问卷的有效回收率均为100%,专家权威系数分别是0.962、0.980,肯德尔协调系数分别是0.220、0.268(均P<0.05),最终形成的中医医院新入职本科护士培训考核评价指标包括5项一级指标、24项二级指标和77项三级指标。结论 构建的中医医院新入职本科护士培训考核评价指标具有较高的可靠性和科学性,可为中医医院管理者及培训者选拔人才提供参考。

农业知识驱动服务技术革新综述与前沿

农业知识驱动服务技术是指运用先进信息技术,科学、高效调配农业领域专业知识服务资源,为农业行业提供智能化知识服务的技术,在解决农业技术服务供需严重失衡等难点问题方面具有重要意义,日益成为支撑农业转型升级和高质量发展的重要引擎,代表着核心研究方向,伴随着技术发展全过程。目前农业行业迫切需要解决的是知识供给严重不足、服务效率不高的问题。农业知识驱动服务技术经历较长时间发展,在知识高效匹配和精准供给方面取得了较大进步,特别是2022年11月以来ChatGPT这类技术的出现,充分展现了超大规模预训练模型在知识智能服务方面的巨大潜力,这也是农业知识驱动服务可以取得突破的关键所在,可以在这方面发挥重要作用。该文在分析农业知识驱动服务相关技术现状的基础上,展望了农业领域可行的知识驱动服务技术路径,预测农业领域知识服务大模型研发构建会呈现参数由少到多、算力由弱趋强、强化训练逐渐加深的特点得到快速发展应用,未来将在专业技术指导、农业“装备-信息-农艺”融合、农业信息系统平台服务总线等方面系统升级现有农业知识服务范式,多模态服务将得到系统融合加深,人机交互模式将向“人性化”方向进一步黏合增强,从而为农业智能化转型升级提供全新的技术支撑,引领农业知识服务从数据检索、语义匹配迈向生成式知识驱动模式转变。

汽车故障知识图谱构建及应用研究

知识图谱技术对汽车高效的故障诊断具有重要的意义,现有汽车故障知识图谱构建存在着实体识别模型效果不佳、无法解决嵌套实体等问题。针对上述问题,通过采用全词掩码的预训练语义模型、加入对抗训练和改进嵌套实体识别模型的方式提高实体识别模型效果,提出了一种改进的嵌套实体识别模型。实验结果表明,所提模型F1值(F_(1))、精确率(P)和召回率(R)相比基线模型分别提高了3.56%、4.08%、3.05%,相比其他模型也有不同程度的提高,验证了所提模型对汽车维修领域实体识别具有显著效果。同时,基于构建的汽车故障知识图谱,实现了汽车故障知识智能问答原型系统,展示了知识图谱技术在汽车故障诊断与维护领域的应用前景。

基于客观结构化临床考试的情景模拟式教学在手术室专科护士培训中的应用

目的探讨基于客观结构化临床考试(objective structured clinical examination,OSCE)的情景模拟式教学在手术室专科护士培训中的应用效果。方法选取参加郑州大学第一附属医院手术室专科护士培训的78名学员为研究对象。按照培训时间先后顺序,将2021年4月-6月参加培训的40名学员作为对照组,将2022年4月-6月参加培训的38名学员作为观察组。对照组采用常规情景模拟式教学,观察组采用基于OSCE的情景模拟式教学,比较2组操作考核成绩、临床思维能力。结果观察组操作考核中患者核查交接、手术体位安置、职业暴露防护的考核成绩均好于对照组(t=4.616,P<0.001;t=3.004,P=0.004;t=3.574,P=0.001),观察组临床思维能力得分高于对照组(t=7.309,P<0.001)。结论将基于OSCE的情景模拟式教学应用于手术室专科护士培训,有助于提高其操作技能和临床思维能力,值得临床推广应用。

增强提示学习的少样本文本分类方法

针对少样本文本分类任务,提出提示学习增强的分类算法(EPL4FTC)。该算法将文本分类任务转换成基于自然语言推理的提示学习形式,在利用预训练语言模型先验知识的基础上实现隐式数据增强,并通过两种粒度的损失进行优化。为捕获下游任务中含有的类别信息,采用三元组损失联合优化方法,并引入掩码语言模型任务作为正则项,提升模型的泛化能力。在公开的4个中文文本和3个英文文本分类数据集上进行实验评估,结果表明EPL4FTC方法的准确度明显优于所对比的基线方法。

基于BERT-BiLSTM-CRF模型的畜禽疫病文本分词研究

针对畜禽疫病文本语料匮乏、文本内包含大量疫病名称及短语等未登录词问题,提出了一种结合词典匹配的BERT-BiLSTM-CRF畜禽疫病文本分词模型。以羊疫病为研究对象,构建了常见疫病文本数据集,将其与通用语料PKU结合,利用BERT(Bidirectional encoder representation from transformers)预训练语言模型进行文本向量化表示;通过双向长短时记忆网络(Bidirectional long short-term memory network,BiLSTM)获取上下文语义特征;由条件随机场(Conditional random field,CRF)输出全局最优标签序列。基于此,在CRF层后加入畜禽疫病领域词典进行分词匹配修正,减少在分词过程中出现的疫病名称及短语等造成的歧义切分,进一步提高了分词准确率。实验结果表明,结合词典匹配的BERT-BiLSTM-CRF模型在羊常见疫病文本数据集上的F1值为96.38%,与jieba分词器、BiLSTM-Softmax模型、BiLSTM-CRF模型、未结合词典匹配的本文模型相比,分别提升11.01、10.62、8.3、0.72个百分点,验证了方法的有效性。与单一语料相比,通用语料PKU和羊常见疫病文本数据集结合的混合语料,能够同时对畜禽疫病专业术语及疫病文本中常用词进行准确切分,在通用语料及疫病文本数据集上F1值都达到95%以上,具有较好的模型泛化能力。该方法可用于畜禽疫病文本分词。