目的分析宁夏回族自治区儿童青少年近视流行现状、影响因素及不同学段间的差异。方法采用分层整群随机抽样的方法,于2019年9月至12月,在宁夏回族自治区银川市、吴忠市、石嘴山市、固原市和中卫市,随机抽取8所小学、6所初中、6所高中、4...目的分析宁夏回族自治区儿童青少年近视流行现状、影响因素及不同学段间的差异。方法采用分层整群随机抽样的方法,于2019年9月至12月,在宁夏回族自治区银川市、吴忠市、石嘴山市、固原市和中卫市,随机抽取8所小学、6所初中、6所高中、4所大学的学生为研究对象,小学每个年级抽取5个班级,初中至大学每个年级抽取4个班级,以抽取班级的全体学生作为研究对象,共抽取学生14211人,对其进行问卷调查、体格检查和视力测量。不同学段儿童近视的影响因素采用最小绝对收缩和选择算子(LASSO)联合Logistic回归进行分析,选择贝叶斯信息准则(Bayesian information criterion,BIC)最小的模型为最优模型。结果宁夏回族自治区儿童青少年近视检出率为70.3%,女生高于男生,城市高于乡镇,差异均有统计学意义(均为P<0.001);按学段分层后,随着年级的增加,近视检出率随之升高,小学最低,大学最高,不同学段近视检出率差异有统计学意义(P<0.001)。近视影响因素的LASSO-Logistic回归分析表明,城乡、性别、年龄、目前是否配戴眼镜、每日课间操节数、是否积极参加体力活动和过去6个月是否保持规律活动是小学生近视的影响因素(均为P<0.05);性别、目前是否配戴眼镜是初中生和高中生近视的影响因素(均为P<0.05);目前是否配戴眼镜是大学生近视的影响因素(P<0.05)。结论宁夏回族自治区儿童青少年近视检出率高,不同学段儿童青少年近视影响因素差异明显。配戴眼镜是控制近视的保护因素。应根据儿童青少年所处学段开展有针对性的视力相关知识的健康教育,增强其健康保健意识,提高儿童青少年视力。展开更多
功能超网络广泛地应用于脑疾病诊断和分类研究中,而现有的关于超网络创建的研究缺乏解释分组效应的能力或者仅考虑到脑区间组级的信息,这样构建的脑功能超网络会丢失一些有用的连接或包含一些虚假的信息,因此,考虑到脑区间的组结构问题...功能超网络广泛地应用于脑疾病诊断和分类研究中,而现有的关于超网络创建的研究缺乏解释分组效应的能力或者仅考虑到脑区间组级的信息,这样构建的脑功能超网络会丢失一些有用的连接或包含一些虚假的信息,因此,考虑到脑区间的组结构问题,引入sparse group Lasso(sgLasso)方法进一步改善超网络的创建。首先,利用sgLasso方法进行超网络创建;然后,引入两组超网络特有的属性指标进行特征提取以及特征选择,这些指标分别是基于单一节点的聚类系数和基于一对节点的聚类系数;最后,将特征选择后得到的两组有显著差异的特征通过多核学习进行特征融合和分类。实验结果表明,所提方法经过多特征融合取得了87.88%的分类准确率。该结果表明为了改善脑功能超网络的创建,需要考虑到组信息,但不能逼迫使用整组信息,可以适当地对组结构进行扩展。展开更多
This study is intended to explore the chemical differences of Acori Tatarinowii Rhizoma （ATR） samples collected from two habitats, Sichuan and Anhui provinces, China. Gas chromatography-mass spectrometry （GC-MS） w...This study is intended to explore the chemical differences of Acori Tatarinowii Rhizoma （ATR） samples collected from two habitats, Sichuan and Anhui provinces, China. Gas chromatography-mass spectrometry （GC-MS） was applied to establishing the quantitative chemical fingerprints of ATRs. A total of 104 volatile compounds were identified and quantified with the information of mass spectra and retention index （RI）. Furthermore, least absolute shrinkage and selection operator （LASSO）, a sparse regularization method, combined with subsampling was employed to improve the classification ability of partial least squares-discriminant analysis （PLS-DA）. After variable selection by LASSO, three chemical markers,β-elemene, α-selinene and α-asarone, were identified for the discrimination of ATRs from two habitats, and the total classification correct rate was increased from 82.76% to 96.55%. The proposed LASSO-PLS-DA method can serve as an efficient strategy for screening marked chemical components and geo-herbalism research of traditional Chinese medicines.展开更多
Fluorescence molecular tomography(FMT)is a fast-developing optical imaging modalitythat has great potential in early diagnosis of disease and drugs development.However,recon-struction algorithms have to address a high...Fluorescence molecular tomography(FMT)is a fast-developing optical imaging modalitythat has great potential in early diagnosis of disease and drugs development.However,recon-struction algorithms have to address a highly ill-posed problem to fulfll 3D reconstruction inFMT.In this contribution,we propose an efficient iterative algorithm to solve the large-scalereconstruction problem,in which the sparsity of fluorescent targets is taken as useful a prioriinformation in designing the reconstruction algorithm.In the implementation,a fast sparseapproximation scheme combined with a stage-wise learning strategy enable the algorithm to dealwith the ill-posed inverse problem at reduced computational costs.We validate the proposed fastiterative method with numerical simulation on a digital mouse model.Experimental results demonstrate that our method is robust for different finite element meshes and different Poissonnoise levels.展开更多
Least Absolute Shrinkage and Selection Operator (LASSO) is used for variable selection as well as for handling the multicollinearity problem simultaneously in the linear regression model. LASSO produces estimates havi...Least Absolute Shrinkage and Selection Operator (LASSO) is used for variable selection as well as for handling the multicollinearity problem simultaneously in the linear regression model. LASSO produces estimates having high variance if the number of predictors is higher than the number of observations and if high multicollinearity exists among the predictor variables. To handle this problem, Elastic Net (ENet) estimator was introduced by combining LASSO and Ridge estimator (RE). The solutions of LASSO and ENet have been obtained using Least Angle Regression (LARS) and LARS-EN algorithms, respectively. In this article, we proposed an alternative algorithm to overcome the issues in LASSO that can be combined LASSO with other exiting biased estimators namely Almost Unbiased Ridge Estimator (AURE), Liu Estimator (LE), Almost Unbiased Liu Estimator (AULE), Principal Component Regression Estimator (PCRE), r-k class estimator and r-d class estimator. Further, we examine the performance of the proposed algorithm using a Monte-Carlo simulation study and real-world examples. The results showed that the LARS-rk and LARS-rd algorithms,?which are combined LASSO with r-k class estimator and r-d class estimator,?outperformed other algorithms under the moderated and severe multicollinearity.展开更多
基于粒计算视角,提出粒化-融合框架下的海量高维数据特征选择算法.运用BLB(Bag of Little Bootstrap)的思想,首先将原始海量数据集粒化为小规模数据子集(粒),然后在每个粒上构建多个自助子集的套索模型,实现粒特征选择,最后,各粒特征选...基于粒计算视角,提出粒化-融合框架下的海量高维数据特征选择算法.运用BLB(Bag of Little Bootstrap)的思想,首先将原始海量数据集粒化为小规模数据子集(粒),然后在每个粒上构建多个自助子集的套索模型,实现粒特征选择,最后,各粒特征选择结果按权重融合、排序,得到原始数据集的有序特征选择结果.人工数据集和真实数据集上的实验表明文中算法对海量高维数据集进行特征选择的可行性和有效性.展开更多
文摘目的分析宁夏回族自治区儿童青少年近视流行现状、影响因素及不同学段间的差异。方法采用分层整群随机抽样的方法,于2019年9月至12月,在宁夏回族自治区银川市、吴忠市、石嘴山市、固原市和中卫市,随机抽取8所小学、6所初中、6所高中、4所大学的学生为研究对象,小学每个年级抽取5个班级,初中至大学每个年级抽取4个班级,以抽取班级的全体学生作为研究对象,共抽取学生14211人,对其进行问卷调查、体格检查和视力测量。不同学段儿童近视的影响因素采用最小绝对收缩和选择算子(LASSO)联合Logistic回归进行分析,选择贝叶斯信息准则(Bayesian information criterion,BIC)最小的模型为最优模型。结果宁夏回族自治区儿童青少年近视检出率为70.3%,女生高于男生,城市高于乡镇,差异均有统计学意义(均为P<0.001);按学段分层后,随着年级的增加,近视检出率随之升高,小学最低,大学最高,不同学段近视检出率差异有统计学意义(P<0.001)。近视影响因素的LASSO-Logistic回归分析表明,城乡、性别、年龄、目前是否配戴眼镜、每日课间操节数、是否积极参加体力活动和过去6个月是否保持规律活动是小学生近视的影响因素(均为P<0.05);性别、目前是否配戴眼镜是初中生和高中生近视的影响因素(均为P<0.05);目前是否配戴眼镜是大学生近视的影响因素(P<0.05)。结论宁夏回族自治区儿童青少年近视检出率高,不同学段儿童青少年近视影响因素差异明显。配戴眼镜是控制近视的保护因素。应根据儿童青少年所处学段开展有针对性的视力相关知识的健康教育,增强其健康保健意识,提高儿童青少年视力。
文摘功能超网络广泛地应用于脑疾病诊断和分类研究中,而现有的关于超网络创建的研究缺乏解释分组效应的能力或者仅考虑到脑区间组级的信息,这样构建的脑功能超网络会丢失一些有用的连接或包含一些虚假的信息,因此,考虑到脑区间的组结构问题,引入sparse group Lasso(sgLasso)方法进一步改善超网络的创建。首先,利用sgLasso方法进行超网络创建;然后,引入两组超网络特有的属性指标进行特征提取以及特征选择,这些指标分别是基于单一节点的聚类系数和基于一对节点的聚类系数;最后,将特征选择后得到的两组有显著差异的特征通过多核学习进行特征融合和分类。实验结果表明,所提方法经过多特征融合取得了87.88%的分类准确率。该结果表明为了改善脑功能超网络的创建,需要考虑到组信息,但不能逼迫使用整组信息,可以适当地对组结构进行扩展。
基金Project(21465016)supported by the National Natural Foundation of China
文摘This study is intended to explore the chemical differences of Acori Tatarinowii Rhizoma （ATR） samples collected from two habitats, Sichuan and Anhui provinces, China. Gas chromatography-mass spectrometry （GC-MS） was applied to establishing the quantitative chemical fingerprints of ATRs. A total of 104 volatile compounds were identified and quantified with the information of mass spectra and retention index （RI）. Furthermore, least absolute shrinkage and selection operator （LASSO）, a sparse regularization method, combined with subsampling was employed to improve the classification ability of partial least squares-discriminant analysis （PLS-DA）. After variable selection by LASSO, three chemical markers,β-elemene, α-selinene and α-asarone, were identified for the discrimination of ATRs from two habitats, and the total classification correct rate was increased from 82.76% to 96.55%. The proposed LASSO-PLS-DA method can serve as an efficient strategy for screening marked chemical components and geo-herbalism research of traditional Chinese medicines.
基金supported by the National Natural Science Foundation of China(Grant No.61372046)the Research Fund for the Doctoral Program ofHigher Education of China(New Teachers)(Grant No.20116101120018)+4 种基金the China Postdoctoral Sci-ence_Foundation_Funded Project(Grant_Nos.2011M501467 and 2012T50814)the Natural Sci-ence Basic Research Plan in Shaanxi Province of China(Grant No.2011JQ1006)the Fund amental Research Funds for the Central Universities(Grant No.GK201302007)Science and Technology Plan Program in Shaanxi Province of China(Grant Nos.2012 KJXX-29 and 2013K12-20-12)the Scienceand Technology Plan Program in Xi'an of China(Grant No.CXY 1348(2)).
文摘Fluorescence molecular tomography(FMT)is a fast-developing optical imaging modalitythat has great potential in early diagnosis of disease and drugs development.However,recon-struction algorithms have to address a highly ill-posed problem to fulfll 3D reconstruction inFMT.In this contribution,we propose an efficient iterative algorithm to solve the large-scalereconstruction problem,in which the sparsity of fluorescent targets is taken as useful a prioriinformation in designing the reconstruction algorithm.In the implementation,a fast sparseapproximation scheme combined with a stage-wise learning strategy enable the algorithm to dealwith the ill-posed inverse problem at reduced computational costs.We validate the proposed fastiterative method with numerical simulation on a digital mouse model.Experimental results demonstrate that our method is robust for different finite element meshes and different Poissonnoise levels.
文摘Least Absolute Shrinkage and Selection Operator (LASSO) is used for variable selection as well as for handling the multicollinearity problem simultaneously in the linear regression model. LASSO produces estimates having high variance if the number of predictors is higher than the number of observations and if high multicollinearity exists among the predictor variables. To handle this problem, Elastic Net (ENet) estimator was introduced by combining LASSO and Ridge estimator (RE). The solutions of LASSO and ENet have been obtained using Least Angle Regression (LARS) and LARS-EN algorithms, respectively. In this article, we proposed an alternative algorithm to overcome the issues in LASSO that can be combined LASSO with other exiting biased estimators namely Almost Unbiased Ridge Estimator (AURE), Liu Estimator (LE), Almost Unbiased Liu Estimator (AULE), Principal Component Regression Estimator (PCRE), r-k class estimator and r-d class estimator. Further, we examine the performance of the proposed algorithm using a Monte-Carlo simulation study and real-world examples. The results showed that the LARS-rk and LARS-rd algorithms,?which are combined LASSO with r-k class estimator and r-d class estimator,?outperformed other algorithms under the moderated and severe multicollinearity.
文摘基于粒计算视角,提出粒化-融合框架下的海量高维数据特征选择算法.运用BLB(Bag of Little Bootstrap)的思想,首先将原始海量数据集粒化为小规模数据子集(粒),然后在每个粒上构建多个自助子集的套索模型,实现粒特征选择,最后,各粒特征选择结果按权重融合、排序,得到原始数据集的有序特征选择结果.人工数据集和真实数据集上的实验表明文中算法对海量高维数据集进行特征选择的可行性和有效性.
