Multiple regression analysis of risk factors related to radiation pneumonitis

BACKGROUND Radiation pneumonitis(RP)is a severe complication of thoracic radiotherapy that may lead to dyspnea and lung fibrosis,and negatively affects patients’quality of life.AIM To carry out multiple regression analysis on the influencing factors of radiation pneumonitis.METHODS Records of 234 patients receiving chest radiotherapy in Huzhou Central Hospital(Huzhou,Zhejiang Province,China)from January 2018 to February 2021,and the patients were divided into either a study group or a control group based on the presence of radiation pneumonitis or not.Among them,93 patients with radiation pneumonitis were included in the study group and 141 without radiation pneumonitis were included in the control group.General characteristics,and radiation and imaging examination data of the two groups were collected and compared.Due to the statistical significance observed,multiple regression analysis was performed on age,tumor type,chemotherapy history,forced vital capacity(FVC),forced expiratory volume in the first second(FEV1),carbon monoxide diffusion volume(DLCO),FEV1/FVC ratio,planned target area(PTV),mean lung dose(MLD),total number of radiation fields,percentage of lung tissue in total lung volume(vdose),probability of normal tissue complications(NTCP),and other factors.RESULTS The proportions of patients aged≥60 years and those with the diagnosis of lung cancer and a history of chemotherapy in the study group were higher than those in the control group(P<0.05);FEV1,DLCO,and FEV1/FVC ratio in the study group were lower than those in the control group(P<0.05),while PTV,MLD,total field number,vdose,and NTCP were higher than in the control group(P<0.05).Logistic regression analysis showed that age,lung cancer diagnosis,chemotherapy history,FEV1,FEV1/FVC ratio,PTV,MLD,total number of radiation fields,vdose,and NTCP were risk factors for radiation pneumonitis.CONCLUSION We have identified patient age,type of lung cancer,history of chemotherapy,lung function,and radiotherapy parameters as risk factors for radiation pneumonitis.Comprehensive evaluation and examination should be carried out before radiotherapy to effectively prevent radiation pneumonitis.

COX MULTIVARIATE REGRESSION ANALYSIS OF RECURRENCE FACTORS FOR COLONIC CARCINOMA

Objective: To determine the independent prognostic factors in the recurrence of colonic carcinoma after curative resection. Methods: Two hundred and one patients undergoing curative resections for colonic carcinoma were investigated by univariate and Cox multivariate regression analyses. Ten factors contributed to the rate were analyzed. Results: Dukes stages, obstruction, postoperative chemotherapy as well as the growth manner of the tumor were significantly associated with the recurrence rate of colonic carcinoma (P<0.05) by univariate analysis, while Dukes stages, obstruction, and postoperative chemotherapy were significant factors by the multivariate analysis. Conclusion: Dukes stages, obstruction, and postoperative chemotherapy are independent prognostic factors in the recurrence of colonic carcinoma.

Logistic Regression Analysis of Influencing Factors on Serum ALT and HCV RNA Changes in Patients with Chronic Hepatitis C

Objective This study was undertaken to investigate the influencing factors on serum ALT level and hepatitis C virus(HCV)RNA titer in chronic hepatitis C(CHC)patients.Methods All patients enrolled into this study were anti-HCV positive.Retrospective tracing method was applied to detect serum ALT level and HCV RNA titer and to collect general information of the patients such as genders,age groups,interferon medication history,infection pathways,height and weight.Then the multi-factor analysis was adopted with the application of binominal logistic regression mode.Results The abnormal rate of ALT level was positively correlated to HCV RNA and gender while negatively correlated to interferon medication history and age group,with Wald value of the 4 factors as 39.604,11.823,18.991 and 7.389,respectively.The positive rate of HCV RNA was negatively correlated to interferon medication history and gender while positively correlated to ALT level,with corresponding Wald value of the 3 factors as81.394,7.618 and 27.562,respectively.Conclusions The normal ALT level in HCV infected patients was associated with viral load,age,gender and interferon medication history,while the normal rate of HCV RNA titer was closely associated with gender,interferon medication history and ALT level.

Logistic Regression Analysis the Risk Factors of Peripherally Inserted Central Catheter Related Blood Stream Infection of Tumor Patients

Objective: Our object is to study risk factors of tumor patients’ PICC catheter-related blood stream infection. Method: a retrospective analysis of data of 586 PICC catheterized patients was implemented, a univariate analysis of general data and catheterizing data of tumor patients was then carried out, and data of single factors with statistical significance were incorporated into multi-factor Logistic regression model for analysis. Results: PICC catheter-related blood stream infection occurred to 16 patients, and occurrence rate was 2.73%. Multi-factor Logistic regression analysis results showed that number of puncturing times, positioning method and maintenance frequency were risk factors for tumor patients’ peripherally inserted central catheter catheter-related blood stream infection, and odds risk values were respectively 8.762, 9.253 and 10.324. Conclusion: for tumor patients implanted with peripherally inserted central catheters, using ECG positioning during strict sterile operation and catheterizing process to avoid repeated puncturing and increasing maintenance frequency could effectively reduce occurrence of PICC catheter-related blood stream infection.

Risk factor analysis of post-ERCP cholangitis: A single-center experience

Background: Endoscopic retrograde cholangiopancreatography(ERCP) may have complications. Our study aimed to investigate the risk factors and prevention of post-ERCP cholangitis.Methods: We retrospectively analyzed 4234 cases undergone ERCP in the Affiliated Drum Tower Hospital of Nanjing University Medical School from January 2008 to December 2013. Patient-related factors and procedure-related factors were analyzed to find the risk factors of post-ERCP cholangitis. The time point of post-ERCP cholangitis was also analyzed. Univariate and multivariate analyses were performed to define the independent risk factors of post-ERCP cholangitis.Results: The success rate of ERCP was 96.8%(4099/4234). The overall complication rate was 9.4%(399/4234). Post-ERCP cholangitis occurred in 102 cases(2.4%, 102/4234). The most dangerous time of post-ERCP cholangitis was from 24 h–48 h after ERCP(45.1%, 46/102). Univariate analysis revealed that age, hypertension, diabetes, previous ERCP history, biliary stent insertion, pancreatography, endoscopic sphincterotomy, balloon dilation and hilar obstruction were risk factors of post-ERCP cholangitis(P < 0.05). Multivariate analysis indicated that age, previous ERCP history and hilar obstruction were independent risk factors(P < 0.05). While endoscopic stone extraction was the potential protective factor.Conclusions: Many risk factors are involved in post-ERCP cholangitis. Among them, old age, previous ERCP history and hilar obstruction were independently related to this post-ERCP complication.

The Stochastic Regression Analysis as a Tool in Ecotoxicological QSAR Studies

Correspondence factor analysis(CFA)was used in conjunction with linear regression analysis to examine the structure-activity relationships of 50 benzene derivatives tested on Pimephales promelas.From nine molecular descriptors(numbers of C,H,O,N,Br,Cl,NO_2,OH,and NH_2 included in the molecules),CFA made it possible to define five new independent variables which were introduced in a stepwise regression analysis procedure to describe the acute toxicity(96-h LC_(50))of the aromatic compounds.The model log 1/C=-0.727F_1+1.248F_3+4.052(r=0.918; s=0.270)is more relevant to describe the ecotoxicological behavior of the studied compounds on the fathead minnow than that obtained with principal components(log 1/C=0.151 PC_1 -0.271 PC_2+4.124;r=0.737;s=0.460).The heuristic potency of this particular statistical analysis,which is called stochastic regression analysis,is discussed in detail.1989 Academic Press,Inc.

Turning Research into Action: Using Factor Analysis to Enhance Program Evaluation

Purpose: Sexual activity among adolescents in the United States remains high. Nearly 46% of students grades 9-12 have engaged in sexual intercourse. One of the more recent statistical tools employed in evaluation efforts includes factor analysis. The objective of this study was to investigate the underlying dimensions of a survey instrument that assesses a youth character development program, which focuses on avoiding high-risk behaviors. Method: The 76-item survey instrument was administered to adolescents (age 12-18). During the 2009-2010 school year, 652 participants in the intervention group and 1110 participants in the comparison group completed the pre-, post-, and 6-month follow-up survey. Results: Using Ajzen’s theory of planned behavior groupings, 27 survey items were selected. Through iterative principal axis factoring, four factors were extracted and rotated. A visual scree plot was generated to determine the number of acceptable factors. The extracted factors accounted for 52.53% of the total variance. Factors were subjected to Equimax rotation with Kaiser normalization and converged after six iterations. Variables with patterned weights less than 0.44 were excluded. A reliability analysis demonstrated internal consistency. Conclusions: Identified factors included: 1) Teenagers’ attitudes toward sexual health behaviors;2) Teenagers’ perceptions of the consequences of sexual health behaviors;3) Parental or guardian expectations;and 4) Teenagers’ relationships with parents or guardians. This study’s results indicated that all factors can be described within Ajzen’s theoretical framework consistent with previous research findings. Results may be used to enhance delivery of the intervention.

An Empirical Study on the Influence of Agricultural Public Infrastructure on Total Factor Productivity of China’s Grain Production

A critical method of ensuring grain production is to increase the total factor productivity(TFP),and the key measure to increase the TFP of grain production lies in the construction of agricultural public infrastructure.For this topic,existing literature lacks systematic and empirical analysis.Therefore,research on the influence of agricultural public infrastructure on the TFP of China’s grain production has relatively strong policy implications and theoretical value.For this study,we collected panel data for grain inputs and outputs as well as for agricultural public infrastructure in China’s provinces(autonomous regions/municipalities)from 1990 to 2017,and adopted the stochastic frontier function(SFF)approach to measure the TFP of provincial-level grain production.Through this empirical study,we analyzed the influence of agricultural public infrastructures,such as irrigation,roads,and electric power facilities on the TFP of China’s agriculture.We found that such facilities have a positive influence on the TFP of grain production.Specifically,when the input for irrigation facilities is increased by 1 percent,the TFP of grain production will rise by 5.74 percent.Based on this finding,policy recommendations are proposed for enhancing grain TFP through agricultural public infrastructure construction.

Incorporating empirical knowledge into data-driven variable selection for quantitative analysis of coal ash content by laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy(LIBS)has become a widely used atomic spectroscopic technique for rapid coal analysis.However,the vast amount of spectral information in LIBS contains signal uncertainty,which can affect its quantification performance.In this work,we propose a hybrid variable selection method to improve the performance of LIBS quantification.Important variables are first identified using Pearson's correlation coefficient,mutual information,least absolute shrinkage and selection operator(LASSO)and random forest,and then filtered and combined with empirical variables related to fingerprint elements of coal ash content.Subsequently,these variables are fed into a partial least squares regression(PLSR).Additionally,in some models,certain variables unrelated to ash content are removed manually to study the impact of variable deselection on model performance.The proposed hybrid strategy was tested on three LIBS datasets for quantitative analysis of coal ash content and compared with the corresponding data-driven baseline method.It is significantly better than the variable selection only method based on empirical knowledge and in most cases outperforms the baseline method.The results showed that on all three datasets the hybrid strategy for variable selection combining empirical knowledge and data-driven algorithms achieved the lowest root mean square error of prediction(RMSEP)values of 1.605,3.478 and 1.647,respectively,which were significantly lower than those obtained from multiple linear regression using only 12 empirical variables,which are 1.959,3.718 and 2.181,respectively.The LASSO-PLSR model with empirical support and 20 selected variables exhibited a significantly improved performance after variable deselection,with RMSEP values dropping from 1.635,3.962 and 1.647 to 1.483,3.086 and 1.567,respectively.Such results demonstrate that using empirical knowledge as a support for datadriven variable selection can be a viable approach to improve the accuracy and reliability of LIBS quantification.

An Empirical Study of Code Clone Clustering Based on Clone Evolution

There are lots of code clones appearing in software,which are similar code fragments with each other. In the past decades,researchers have proposed some state-of-the-art methods to detect clones. The code clones have showing some relationship with the evolution of software. In order to explore relationships between clones and their evolution,we propose a framework to cluster clones with a Fuzzy C-means clustering method.Firstly,we detect all the clones using Ni Cad,and build the clone genealogies for multiple versions software.Secondly,we extract some metrics to describe the clones and their evolution. Finally,we cluster all clone's vectors,which are generated with the different metrics for different proposes. Experimental results on six open source software packages have shown the relationships among the clone life,the number of change times,the clone pattern and et al. can help developers to understand clones.

Quality of life of hospitalized patients after lung cancer operation and analysis of influencing factors

Objective:To explore the current status and influencing factors of quality of life in patients with lung cancer after surgery in a tertiary hospital in Hainan province.Methods:To investigate the influencing factors of quality of life of lung cancer patients after surgery in a tertiary hospital in Hainan province by cross‑sectional survey method.Results:The scores of insomnia,appetite loss,constipation and pain in 186 lung cancer patients after surgery in a tertiary hospital in Hainan Province were significantly higher than the reference value.Multiple linear regression analysis showed that older patients(>60 years)had lower scores in physical function domain(β=-0.193),and female patients had more appetite loss symptoms(β=0.245).Compared with other minority ethnic groups,Han ethnic group had lower scores in role function domain(β=0.179),more severe fatigue symptoms(β=-0.162),and higher general health level(β=0.166).Patients with employee medical insurance had lower scores of emotional function(β=0.194),cognitive function(β=0.281),the lowest score in social function(β=0.188),and severe pain in other parts(β=-0.227).Smokers had less cough symptoms(β=0.175)and more arm and shoulder pain symptoms(β=-0.21)than non‑smokers.Patients with secondhand smoke exposure had lower cognitive function scores(β=-0.158)and more obvious symptoms of oral ulcer(β=0.185).Patients who drank alcohol frequently(drinking frequency>1 time/day)had more severe cough symptoms(β=0.27).Patients with small number of children(0‑1)had milder cough symptoms(β=0.178).Patients who did not understand the disease had obvious symptoms of arm and shoulder pain(β=0.151).Patients with early pathological stage(stageⅠ‑Ⅱ)had more severe shortness of breath(β=-0.159)and pain(β=-0.181).The symptoms of appetite loss were more obvious in patients living in cities(β=0.192).The symptoms of peripheral neuropathy were more obvious(β=0.174).Patients who often consumed pickulated food had severe pain symptoms(β=-0.219),and pain in other parts was obvious(β=-0.149).Male patients had obvious alopecia symptoms(β=-0.306).Conclusion:Age,ethnicity,residence,type of medical insurance,number of children,pathological stage of lung cancer,smoking,second‑hand smoke exposure,alcohol consumption,and frequent consumption of pickled food were related to the quality of life of lung cancer patients in hospital after surgery.Medical staff and family members should pay attention to the emotional communication of patients during the treatment of lung cancer patients in hospital after surgery.Patients should avoid exposure to smoking,alcohol and second‑hand smoke,and reduce consumption of pickled food.

Comparison of dimension reduction-based logistic regression models for case-control genome-wide association study:principal components analysis vs.partial least squares

With recent advances in biotechnology, genome-wide association study （GWAS） has been widely used to identify genetic variants that underlie human complex diseases and traits. In case-control GWAS, typical statistical strategy is traditional logistical regression （LR） based on single-locus analysis. However, such a single-locus analysis leads to the well-known multiplicity problem, with a risk of inflating type I error and reducing power. Dimension reduction-based techniques, such as principal component-based logistic regression （PC-LR）, partial least squares-based logistic regression （PLS-LR）, have recently gained much attention in the analysis of high dimensional genomic data. However, the perfor- mance of these methods is still not clear, especially in GWAS. We conducted simulations and real data application to compare the type I error and power of PC-LR, PLS-LR and LR applicable to GWAS within a defined single nucleotide polymorphism （SNP） set region. We found that PC-LR and PLS can reasonably control type I error under null hypothesis. On contrast, LR, which is corrected by Bonferroni method, was more conserved in all simulation settings. In particular, we found that PC-LR and PLS-LR had comparable power and they both outperformed LR, especially when the causal SNP was in high linkage disequilibrium with genotyped ones and with a small effective size in simulation. Based on SNP set analysis, we applied all three methods to analyze non-small cell lung cancer GWAS data.

Logistic Regression Analysis of Syndrome Essential Factors in Patients with Unstable Angina Pectoris

Objective: To explore the correlation between common syndrome essential factors and the symptoms and signs of unstable angina (UA). Methods: Eight hundred and fifteen patients with UA confirmed by coronary angiography were identified from several centers. Common syndrome essential factors were selected on the basis of expert experience. The correlations between common syndrome essential factors and symptoms and signs of UA were analyzed using binary logistic regression analysis. Results: The common syndrome essential factors in unstable angina were blood stasis, qi stagnation, phlegm turbidity, heat stagnancy, qi deficiency, yin deficiency, and yang deficiency. Symptoms such as chest pain, hypochondriac distention, ecchymosis, dark orbits, dark and purplish tongue, and tongue with ecchymosis and petechiae were significant diagnostic features of "blood stasis". Aversion to cold and cool limbs, weakness in the waist and knees, and clear abundant urine were significant diagnostic features of "yang deficiency". These results were in accordance with the understanding of traditional clinical Chinese medical practice. Conclusion: This clinical study analyzed the correlations between common syndrome essential factors and the symptoms and signs of unstable angina. The results provide the basis for establishing diagnostic criteria for syndrome essential factors.

Empirical Study on Growth of Evil Forces in Land Requisition and Relocation in City G of Hubei Province Based on Social Network Analysis

Using social network analysis method,this paper made an empirical study on growth of evil forces in land requisition and relocation in City G of Hubei Province. It obtained following results:(i) lawless developers and inefficient public security organs form interested parties of evil forces. Besides,the inward closeness centrality of evil forces is high,manifesting that evil forces independently possess decentralized power of network and have unscrupulous behavior in land requisition and relocation to a certain extent.(ii) Activities of evil forces have complicated spatial correlation and their geographical distribution is uneven,taking on irregular characteristics. In the field of land requisition and relocation,some evil forces are expandi Xng and spreading,while other forces are relatively weak. In conclusion,growth of evil forces comes from premeditation and collaboration of lawless developers,lack of functions and weak attack of public security organs; growth of evil forces has periodic changes,when in power,they will take opportunity to expand,while losing power,they will hide or even disappear.

The Effects of Agricultural Informatization on Agricultural Economic Growth: An Empirical Analysis Based on Regression Model

This article selects some major factors influencing the agricultural economic growth are selected,such as labor,capital input,farmland area,fertilizer input and information input.And it selects some factors to explain information input,such as the number of website ownership,types of books,magazines and newspapers published,the number of telephone ownership per 100 households,the number of home computers ownership per 100 households,farmers' spending on transportation and communication,culture,education,entertainment and services, and the total number of agricultural science and technology service personnel.Using regression model,this article conducts regression analysis of the cross-section data on 31 provinces,autonomous regions and municipalities in 2010.The results show that the building of information infrastructure,the use of means of information,the popularization and promotion of knowledge of agricultural science and technology,play an important role in promoting agricultural economic growth.

Factors contributing to myasthenic crisis after thymectomy-a multivariate analysis study

Objective To evaluate the risk factors of myasthenic crisis (MC) after thymectomy for myasthenia gravis (MG) . Methods A retrospective study was conducted in 84 patients with MG from July 1995 to December 2009. The following factors were evaluated: sex,age,history of preoperative myasthenic crisis,Ossermen classification, steroid hormones use and anticholinesterase drugs before operation,operation methods,operation time,thymoma,

Diabetes mellitus and prostate cancer risk:A mendelian randomization analysis

BACKGROUND Some studies have directed towards an association between diabetes mellitus(DM)and prostate cancer(PCa);however,this specific relationship remains inconclusive.In recent years,Mendelian randomization(MR)has become a widely used analytical method for inferring epidemiological causes.AIM To investigated the potential relationship between DM and PCa using MR.METHODS We downloaded relevant data on"diabetes"and"PCa"from the IEU OpenGWAS project database,performed three different methods to conduct MR,and carried out sensitivity analysis for verification.RESULTS The results indicated that DM was an independent risk factor for PCa.The odds ratio(OR)values obtained using the inverse variance weighted method in this study were as follows:OR=1.018(95%confidence interval:1.004-1.032),P=0.014.CONCLUSION We found that DM could increase the incidence rate of PCa.

Analysis of environmental factors about cerebral stroke

In this paper, the disease of cerebral stroke is studied mainly. According to the basic information of the patient, Linear fitting method is applied to calculate between the number of sick and age by using MATLAB software. According to data of pressure, temperature and humidity, nondimensionalizing by normalization method, linear regressions are analysed between the number of the patient and pressure, temperature, humidity. According to the important feature and the key indicators of cerebral stroke, high-risk groups are proposed to the early warning and intervene.