Laparoscopic skills training has always been crucial for novice surgeons. Readily accessible equipment, aswell as structured training curriculum should be provided to guarantee adequate practice hours and skillprofici...Laparoscopic skills training has always been crucial for novice surgeons. Readily accessible equipment, aswell as structured training curriculum should be provided to guarantee adequate practice hours and skillproficiency. Dry-lab training is typically adopted before animal model surgery, usually comprising ofpurpose-built bulky simulators that is neither accessible nor portable. In this technical note, we designed ahome-made simulator, using two 4 L water jugs as operating space that are communicated inside, plus anobservation hole taped in between to mimic the triangular working space of laparoscopic surgery. Imagingwas achieved via smartphone camera, which was wirelessly connected to a laptop and a projector for realtime display on multiple screens, using built-in multi-screen collaboration software. A self-regulated andproficiency-based training curriculum was adopted. This dry-lab simulator is low-cost, highly portable andeasily replicable for basic laparoscopic skills training for the beginners to intermediate surgeons, whichmay serve as a good way for the standardized residency and specialist training program.展开更多
[目的]通过将网状Meta分析与多指标决策法(MCDM)相结合,构建多维度评价体系,对针刺治疗强直性脊柱炎的优势方案进行筛选,为临床决策提供科学依据。[方法]检索中国知网、万方知识服务平台、维普数据库、SinoMed、PubMed、EMbase、Cochran...[目的]通过将网状Meta分析与多指标决策法(MCDM)相结合,构建多维度评价体系,对针刺治疗强直性脊柱炎的优势方案进行筛选,为临床决策提供科学依据。[方法]检索中国知网、万方知识服务平台、维普数据库、SinoMed、PubMed、EMbase、Cochrane Library、Web of Science 8个数据库截至2025年1月1日收录的有关针刺治疗强直性脊柱炎的随机对照研究(RCTs),采用改良Jadad量表对证据质量进行评价,网状Meta分析计算有效率RR值和巴氏强直性脊柱炎疾病活动性指数(BASDAI)评分的累积排序概率图下面积(SUCRA)值、加权干预次数和改良Jadad评分,构建决策矩阵。使用线性比例法建立标准化决策矩阵。采用熵权法进行指标权重确定和再分配,并运用TOPSIS法、VIKOR法和PROMETHEE法3种决策方法筛选优势方案。[结果]共纳入17项研究,得到14种干预方案和4种评价指标,3种决策结果之间具有高度相似性,通督热针法为针刺治疗强直性脊柱炎的最佳优势方案。[结论]研究通过网状Meta分析与多准则决策方法相结合,筛选出针刺治疗强直性脊柱炎的优势方案,其中通督热针法为最优选择。两种方法的结合应用为针刺优势方案的筛选提供了新思路,为针灸临床决策提供了参考依据。展开更多
高效准确的机器学习模型是高通量筛选优质含能分子的基础,为此设计了完全通过结构式便可输出多项含能材料性质的多输入多输出机器学习模型(MIMO-ML)。基于478个含能分子数据集进行模型构建,通过筛选得到AMID_h、ATTS1are、OB三个优秀描...高效准确的机器学习模型是高通量筛选优质含能分子的基础,为此设计了完全通过结构式便可输出多项含能材料性质的多输入多输出机器学习模型(MIMO-ML)。基于478个含能分子数据集进行模型构建,通过筛选得到AMID_h、ATTS1are、OB三个优秀描述符,并在此基础上建立了自定义描述符集。结果表明,随机森林(RF)及多层感知机(MLP)适合被应用于含能材料性能预测多输出模型构建,输出性质包括爆速D(MAE=256 m/s),热分解温度T_(d)(MAE=34.7℃),撞击感度ln H 50(MAE=0.63)。同时,MLP模型相比于RF模型对于特征数量更敏感,且利用更少的特征可得到与RF精度相似的模型,表明MIMO-ML模型能够快速且准确地识别高性能含能材料,可应用于含能分子的设计与快速筛选。展开更多
基金This study is supported by the 2021 Changhai Hospital Educational Sponsorship Fund(CHPY2021B24,General Program,YC).
文摘Laparoscopic skills training has always been crucial for novice surgeons. Readily accessible equipment, aswell as structured training curriculum should be provided to guarantee adequate practice hours and skillproficiency. Dry-lab training is typically adopted before animal model surgery, usually comprising ofpurpose-built bulky simulators that is neither accessible nor portable. In this technical note, we designed ahome-made simulator, using two 4 L water jugs as operating space that are communicated inside, plus anobservation hole taped in between to mimic the triangular working space of laparoscopic surgery. Imagingwas achieved via smartphone camera, which was wirelessly connected to a laptop and a projector for realtime display on multiple screens, using built-in multi-screen collaboration software. A self-regulated andproficiency-based training curriculum was adopted. This dry-lab simulator is low-cost, highly portable andeasily replicable for basic laparoscopic skills training for the beginners to intermediate surgeons, whichmay serve as a good way for the standardized residency and specialist training program.
文摘[目的]通过将网状Meta分析与多指标决策法(MCDM)相结合,构建多维度评价体系,对针刺治疗强直性脊柱炎的优势方案进行筛选,为临床决策提供科学依据。[方法]检索中国知网、万方知识服务平台、维普数据库、SinoMed、PubMed、EMbase、Cochrane Library、Web of Science 8个数据库截至2025年1月1日收录的有关针刺治疗强直性脊柱炎的随机对照研究(RCTs),采用改良Jadad量表对证据质量进行评价,网状Meta分析计算有效率RR值和巴氏强直性脊柱炎疾病活动性指数(BASDAI)评分的累积排序概率图下面积(SUCRA)值、加权干预次数和改良Jadad评分,构建决策矩阵。使用线性比例法建立标准化决策矩阵。采用熵权法进行指标权重确定和再分配,并运用TOPSIS法、VIKOR法和PROMETHEE法3种决策方法筛选优势方案。[结果]共纳入17项研究,得到14种干预方案和4种评价指标,3种决策结果之间具有高度相似性,通督热针法为针刺治疗强直性脊柱炎的最佳优势方案。[结论]研究通过网状Meta分析与多准则决策方法相结合,筛选出针刺治疗强直性脊柱炎的优势方案,其中通督热针法为最优选择。两种方法的结合应用为针刺优势方案的筛选提供了新思路,为针灸临床决策提供了参考依据。
文摘高效准确的机器学习模型是高通量筛选优质含能分子的基础,为此设计了完全通过结构式便可输出多项含能材料性质的多输入多输出机器学习模型(MIMO-ML)。基于478个含能分子数据集进行模型构建,通过筛选得到AMID_h、ATTS1are、OB三个优秀描述符,并在此基础上建立了自定义描述符集。结果表明,随机森林(RF)及多层感知机(MLP)适合被应用于含能材料性能预测多输出模型构建,输出性质包括爆速D(MAE=256 m/s),热分解温度T_(d)(MAE=34.7℃),撞击感度ln H 50(MAE=0.63)。同时,MLP模型相比于RF模型对于特征数量更敏感,且利用更少的特征可得到与RF精度相似的模型,表明MIMO-ML模型能够快速且准确地识别高性能含能材料,可应用于含能分子的设计与快速筛选。
文摘碳价是碳市场的核心要素,碳价波动受到众多因素及其时滞效应的影响。为精准预测全国碳市场碳排放配额(Chinese emission allowances,CEA)价格,从关联碳市场、经济发展、国外能源、国内能源和人民币汇率五个维度选取结构化影响因素,从经济政策、环境影响和用户意愿三个维度爬取来自百度搜索引擎的非结构化影响因素,然后引入MIV-BP模型筛选主要的影响因素,并基于最大信息系数(maximum information coefficient,MIC)对碳价以及多源影响因素进行时滞估计。在此基础上,构建融合多源信息的碳价时滞组合预测模型MIC-LSTM-BP,并和基准模型LSTM、BP、LSTM-BP以及时滞基准模型MIC-LSTM、MIC-BP、MIC-LSTM-GBDT进行对比分析,以验证新模型的有效性。结果表明,时滞信息的引入有助于提升模型的预测精度;相较于基准模型和时滞基准模型,MICLSTM-BP模型预测CEA价格精度最高,价格波动追随能力最好。
