AIM： To figure out the contributed factors of the hospitalization expenses of senile cataract patients（HECP） and build up an area-specified senile cataract diagnosis related group（DRG） of Shanghai thereby formula...AIM： To figure out the contributed factors of the hospitalization expenses of senile cataract patients（HECP） and build up an area-specified senile cataract diagnosis related group（DRG） of Shanghai thereby formulating the reference range of HECP and providing scientific basis for the fair use and supervision of the health care insurance fund.METHODS： The data was collected from the first page of the medical records of 22 097 hospitalized patients from tertiary hospitals in Shanghai from 2010 to 2012 whose major diagnosis were senile cataract. Firstly, we analyzed the influence factors of HECP using univariate and multivariate analysis. DRG grouping was conducted according to the exhaustive Chi-squared automatic interaction detector（E-CHAID） model, using HECP as target variable. Finally we evaluated the grouping results using non-parametric test such as Kruskal-Wallis H test, RIV, CV, etc.RESULTS： The 6 DRGs were established as well as criterion of HECP, using age, sex, type of surgery and whether complications/comorbidities occurred as the key variables of classification node of senile cataract cases.CONCLUSION： The grouping of senile cataract cases based on E-CHAID algorithm is reasonable. And the criterion of HECP based on DRG can provide a feasible way of management in the fair use and supervision of medical insurance fund.展开更多
Purpose:This study was carried out to uncover the characteristics of information seeking behavior of researchers as faculty/student team members.Design/methodology/approach:An inventory encompassing 6 dimensions of in...Purpose:This study was carried out to uncover the characteristics of information seeking behavior of researchers as faculty/student team members.Design/methodology/approach:An inventory encompassing 6 dimensions of information seeking behavior was developed:Information awareness,information acquisition,information evaluation,information organization and management,information utilization and information ethics.Data was collected on 306 respondents from 52 faculty/student teams in Central South University in China and analyzed using SPSS 18.0 software.Findings:Significant differences were found among researchers with different genders in information awareness and in different academic disciplines in information acquisition and information utilization.The survey shows the characteristics of information seeking behavior of different gender groups and different teams:1) male participants got higher scores in all of the 6 dimensions of information seeking behavior;2) small teams performed best,followed by middle-sized teams and large teams;3) faculty/doctoral student teams possessed better information seeking skills than faculty/master’s student teams or faculty/doctoral and master’s student teams:4) medical teams achieved the highest level in all of the 6 dimensions of information seeking behavior,whereas natural science teams the lowest level.Medical and engineering teams were rated higher than other teams in information acquisition and information utilization.Research limitations:The small population size and doctoral students accounting for only a small portion of the respondents in the sample limit the generalization of our findings.Practical implications:The findings of this study have some implications for research and practice,especially for educational institutions,library science and information literacy training.Originality/value:This paper is the first to describe and analyze the characteristics of information seeking behavior of researchers as faculty/student team members.展开更多
Background:To compare multiset feature selection and feature fusion algorithm-based contrast-enhanced computed tomography(CE-CT)radiomics models for hepatocellular carcinoma(HCC)microvascular invasion(MVI)prediction u...Background:To compare multiset feature selection and feature fusion algorithm-based contrast-enhanced computed tomography(CE-CT)radiomics models for hepatocellular carcinoma(HCC)microvascular invasion(MVI)prediction using traditional radiomics methods,and to analyze inter-set feature relationships to assess mutual corroboration or complementary effects.Methods:In total,172 patients with histopathologically confirmed MVI from a single center were retrospectively analyzed.Handcrafted(HCF)and deep image features(DF)of the tumor volumes in the CE-CT arterial and portal venous phase images were extracted.A combination of feature selection and feature fusion algorithms,including single-view,multi-task,and multi-view feature selection,and Concatenation-,Canonical Correlation Analysis(CCA)-,or Joint and Individual Variation Explained(JIVE)-based feature fusion,were used to build radiomics models.The top-ranked features were analyzed.Results:Multiset feature selection and feature fusion algorithm-based radiomics models exhibited higher area under the receiver operating characteristic curve(AUC)than baseline models.In HCF dataset,the Minimal-redundancy-maximal-relevance(mRMR)&JIVE-based model exhibited superior prediction efficiency over the baseline model,with AUC:0.7136±0.0338 vs.0.6518±0.0394 in Logistic Regression(LR)classifier,and AUC:0.7155±0.0232 vs.0.6563±0.0331 in Support Vector Machine(SVM)classifier.In the DF dataset,the Adaptive-weighting Discriminative Regression for Multi-view Classification(WeightReg)&JIVE-based model exhibited superior prediction efficiency over the baseline model,with AUC:0.7769±0.0266 vs.0.7410±0.0371 in LR classifier,and AUC:0.7840±0.0328 vs.0.7419±0.0364 in SVM classifier.In the DF&HCF dataset,the mRMR&JIVE-and WeightReg&JIVE-based models exhibited superior prediction efficiency over the baseline model in LR classifier,with AUC:0.7742±0.0460,0.7687±0.0269 vs.0.7491±0.0438,respectively.WeightReg&Concatenation and Adaptive-Similarity-based Multi-modality Feature Selection(amtfs)&Concate-nation models exhibited superior prediction efficiency over the baseline model in SVM classifier,with AUC:0.7821±0.0191,0.7796±0.0238 vs.0.7540±0.0426,respectively.All the results were statistically signifi-cant.Mutually correlated and complementary features in the arterial(Ar)and portal venous(PV)phases were helpful for MVI prediction.展开更多
基金Supported by the Key Research and Development Program of Hunan Province(No.2017SK2011)
文摘AIM： To figure out the contributed factors of the hospitalization expenses of senile cataract patients（HECP） and build up an area-specified senile cataract diagnosis related group（DRG） of Shanghai thereby formulating the reference range of HECP and providing scientific basis for the fair use and supervision of the health care insurance fund.METHODS： The data was collected from the first page of the medical records of 22 097 hospitalized patients from tertiary hospitals in Shanghai from 2010 to 2012 whose major diagnosis were senile cataract. Firstly, we analyzed the influence factors of HECP using univariate and multivariate analysis. DRG grouping was conducted according to the exhaustive Chi-squared automatic interaction detector（E-CHAID） model, using HECP as target variable. Finally we evaluated the grouping results using non-parametric test such as Kruskal-Wallis H test, RIV, CV, etc.RESULTS： The 6 DRGs were established as well as criterion of HECP, using age, sex, type of surgery and whether complications/comorbidities occurred as the key variables of classification node of senile cataract cases.CONCLUSION： The grouping of senile cataract cases based on E-CHAID algorithm is reasonable. And the criterion of HECP based on DRG can provide a feasible way of management in the fair use and supervision of medical insurance fund.
基金supported by the National Social Science Foundation of China(Grant No.:11BTQ044)
文摘Purpose:This study was carried out to uncover the characteristics of information seeking behavior of researchers as faculty/student team members.Design/methodology/approach:An inventory encompassing 6 dimensions of information seeking behavior was developed:Information awareness,information acquisition,information evaluation,information organization and management,information utilization and information ethics.Data was collected on 306 respondents from 52 faculty/student teams in Central South University in China and analyzed using SPSS 18.0 software.Findings:Significant differences were found among researchers with different genders in information awareness and in different academic disciplines in information acquisition and information utilization.The survey shows the characteristics of information seeking behavior of different gender groups and different teams:1) male participants got higher scores in all of the 6 dimensions of information seeking behavior;2) small teams performed best,followed by middle-sized teams and large teams;3) faculty/doctoral student teams possessed better information seeking skills than faculty/master’s student teams or faculty/doctoral and master’s student teams:4) medical teams achieved the highest level in all of the 6 dimensions of information seeking behavior,whereas natural science teams the lowest level.Medical and engineering teams were rated higher than other teams in information acquisition and information utilization.Research limitations:The small population size and doctoral students accounting for only a small portion of the respondents in the sample limit the generalization of our findings.Practical implications:The findings of this study have some implications for research and practice,especially for educational institutions,library science and information literacy training.Originality/value:This paper is the first to describe and analyze the characteristics of information seeking behavior of researchers as faculty/student team members.
基金supported by the Natural Science Foundation of Hunan Province[grant number 2021JJ40951]the Research Project of the Hunan Provincial Health Commission[grant numbers B202309018525 and B202307017799]+2 种基金the National Natural Science Foundation of China[grant number 82303133]the China Postdoctoral Science Foundation[grant number 2022M723559]the Medical Science Research Project of the Hebei Provincial Health Commission[grant number 20200197].
文摘Background:To compare multiset feature selection and feature fusion algorithm-based contrast-enhanced computed tomography(CE-CT)radiomics models for hepatocellular carcinoma(HCC)microvascular invasion(MVI)prediction using traditional radiomics methods,and to analyze inter-set feature relationships to assess mutual corroboration or complementary effects.Methods:In total,172 patients with histopathologically confirmed MVI from a single center were retrospectively analyzed.Handcrafted(HCF)and deep image features(DF)of the tumor volumes in the CE-CT arterial and portal venous phase images were extracted.A combination of feature selection and feature fusion algorithms,including single-view,multi-task,and multi-view feature selection,and Concatenation-,Canonical Correlation Analysis(CCA)-,or Joint and Individual Variation Explained(JIVE)-based feature fusion,were used to build radiomics models.The top-ranked features were analyzed.Results:Multiset feature selection and feature fusion algorithm-based radiomics models exhibited higher area under the receiver operating characteristic curve(AUC)than baseline models.In HCF dataset,the Minimal-redundancy-maximal-relevance(mRMR)&JIVE-based model exhibited superior prediction efficiency over the baseline model,with AUC:0.7136±0.0338 vs.0.6518±0.0394 in Logistic Regression(LR)classifier,and AUC:0.7155±0.0232 vs.0.6563±0.0331 in Support Vector Machine(SVM)classifier.In the DF dataset,the Adaptive-weighting Discriminative Regression for Multi-view Classification(WeightReg)&JIVE-based model exhibited superior prediction efficiency over the baseline model,with AUC:0.7769±0.0266 vs.0.7410±0.0371 in LR classifier,and AUC:0.7840±0.0328 vs.0.7419±0.0364 in SVM classifier.In the DF&HCF dataset,the mRMR&JIVE-and WeightReg&JIVE-based models exhibited superior prediction efficiency over the baseline model in LR classifier,with AUC:0.7742±0.0460,0.7687±0.0269 vs.0.7491±0.0438,respectively.WeightReg&Concatenation and Adaptive-Similarity-based Multi-modality Feature Selection(amtfs)&Concate-nation models exhibited superior prediction efficiency over the baseline model in SVM classifier,with AUC:0.7821±0.0191,0.7796±0.0238 vs.0.7540±0.0426,respectively.All the results were statistically signifi-cant.Mutually correlated and complementary features in the arterial(Ar)and portal venous(PV)phases were helpful for MVI prediction.
