Automatic Linear Modeling Research Articles

Hybrid model for ecological vulnerability assessment in Benin

Identifying ecologically fragile areas by assessing ecosystem vulnerability is an essential task in environmental conservation and management. Benin is considered a vulnerable area, and its coastal zone, which is subject to erosion and flooding effects, is particularly vulnerable. This study assessed terrestrial ecosystems in Benin by establishing a hybrid ecological vulnerability index (EVI) for 2016 that combined a composite model based on principal component analysis (PCA) with an additive model based on exposure, sensitivity and adaptation. Using inverse distance weighted (IDW) interpolation, point data were spatially distributed by their geographic significance. The results revealed that the composite system identified more stable and vulnerable areas than the additive system; the two systems identified 48,600 km2 and 36,450 km2 of stable areas, respectively, for a difference of 12,150 km2, and 3,729 km2 and 3,007 km2 of vulnerable areas, for a difference of 722 km2. Using Moran’s I and automatic linear modeling, we improved the accuracy of the established systems. In the composite system, increases of 11,669 km2 in the potentially vulnerable area and 1,083 km2 in the highly vulnerable area were noted in addition to a decrease of 4331 km2 in the potential area; while in the additive system, an increase of 3,970 km2 in the highly vulnerable area was observed. Finally, southern Benin was identified as vulnerable in the composite system, and both northern and southern Benin were identified as vulnerable in the additive system. However, regardless of the system, Littoral Province in southern Benin, was consistently identified as vulnerable, while Donga Province was stable.

Cognitive Processes Underlying Verbal Fluency in Multiple Sclerosis.

Background: Verbal fluency (VF) has been associated with several cognitive functions, but the cognitive processes underlying verbal fluency deficits in Multiple Sclerosis (MS) are controversial. Further knowledge about VF could be useful in clinical practice, because these tasks are brief, applicable, and reliable in MS patients. In this study, we aimed to evaluate the cognitive processes related to VF and to develop machine-learning algorithms to predict those patients with cognitive deficits using only VF-derived scores.Methods: Two hundred participants with MS were enrolled and examined using a comprehensive neuropsychological battery, including semantic and phonemic fluencies. Automatic linear modeling was used to identify the neuropsychological test predictors of VF scores. Furthermore, machine-learning algorithms (support vector machines, random forest) were developed to predict those patients with cognitive deficits using only VF-derived scores.Results: Neuropsychological tests associated with attention-executive functioning, memory, and language were the main predictors of the different fluency scores. However, the importance of memory was greater in semantic fluency and clustering scores, and executive functioning in phonemic fluency and switching. Machine learning algorithms predicted general cognitive impairment and executive dysfunction, with F1-scores over 67–71%.Conclusions: VF was influenced by many other cognitive processes, mainly including attention-executive functioning, episodic memory, and language. Semantic fluency and clustering were more explained by memory function, while phonemic fluency and switching were more related to executive functioning. Our study supports that the multiple cognitive components underlying VF tasks in MS could serve for screening purposes and the detection of executive dysfunction.

Structural, cultural and instructional predictors essential to sustained implementation fidelity in schools: The School-Wide Positive Behavior Support Model (SWPBS)

Optimal intervention effects require adequate implementation fidelity, and high sustained implementation fidelity is hypothesized a pre-requisite for long-term positive intervention effects. However, implementing interventions with high fidelity seems particularly difficult in schools. The purpose of the this prospective longitudinal study was to add to the current knowledge by testing the predictive effects of school structural, cultural, instructional and community-level factors to longer-term fidelity in a sample of 65 representative primary schools in Norway (Grades 1–7) implementing the evidence-based School-Wide Positive Behavior Support model (SWPBS). Results from the Automatic Linear Modeling analysis which included more than 300 predictor variables, showed that the 10 most important predictors were all related to the schools inner context, in particular to the cultural characteristics of the schools, teachers' disciplinary practices, and prior fidelity. The findings are discussed in relation to prior and future research, and practical implications are coined.

Development, Spanish Normative Data, and Validation of a Social Cognition Battery in Prodromal Alzheimer's Disease and Multiple Sclerosis.

The assessment of social cognition changes may be challenging, especially in the earliest stages of some neurodegenerative diseases. Our objective was to validate a social cognition battery from a multidomain perspective. In this regard, we aimed to adapt several tests, collect normative data, and validate them in prodromal Alzheimer's disease (AD) and multiple sclerosis (MS). A total of 92 healthy controls, 25 prodromal AD, and 39 MS patients were enrolled. Age-, gender-, and education-matched control groups were created for comparisons. Social cognition battery was composed of an emotion-labeling task developed from FACES database, the Story-based Empathy test (SET), the Faux Pas test, and the Interpersonal Reactivity Index. Patients were also evaluated with a comprehensive cognitive battery to evaluate the other cognitive domains. Automatic linear modeling was used to predict each social cognition test's performance using the neuropsychological tests examining other cognitive domains. The reliability of the battery was moderate-high. Significant intergroup differences were found with medium-large effect sizes. Moderate correlations were found between social cognition battery and neuropsychological tests. The emotion labeling task and SET showed moderate correlations with age and education, and age, respectively. Regression-based norms were created considering the relevant demographic variables. Linear regression models including other neuropsychological tests explained between 7.7% and 68.8% of the variance of the social cognition tests performance. Our study provides a battery for the assessment of social cognition in prodromal AD and MS with Spanish normative data to improve the evaluation in clinical and research settings.

Platelet-to-lymphocyte ratio is correlated with a delay in feeding resumption following a transhiatal esophagectomy with cervical anastomosis

IntroductionThe lymphocytic population, neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) are prognostic tools predictive of adverse outcomes for several solid tumors and oncologic surgeries, one of which is esophageal adenocarcinoma. Furthermore, delayed resumption of oral feeding postoperatively is associated with significant morbidity. Given the controversies regarding post-op nutritional support in these patients, this study investigates the prognostic role of the lymphocytic percentage, the NLR, and the PLR in predicting prolonged length of hospital stay (LOHS) and ICU stay (LOICUS) as well as delayed oral feeding following transhiatal esophagectomy (THE) for adenocarcinoma of the esophagogastric junction (AEG).MethodsForty consecutive patients who underwent transhiatal esophagectomy performed by a single surgeon for Siewert type II and type III adenocarcinoma of the esophagogastric junction at a tertiary referral center were selected. Retrospective data collection was performed from the patients’ medical records, and statistical analysis was performed using Pearson correlation and Student’s t test and Chi-square testing.ResultsAn increased LOHS was correlated with a lower preoperative lymphocyte percentage (p = 0.043), higher NLR (p = 0.010) and PLR (p = 0.015), and an increased number of packed red blood cell (PRBC) transfusions perioperatively (p = 0.030). An increased LOICUS was correlated with a lower preoperative lymphocyte percentage (p = 0.033), higher NLR (p = 0.018) and PLR (p = 0.044), an increased number of PRBC transfusions (p = 0.001), and patients’ comorbidities (p < 0.05). A delay in feeding resumption was correlated with a lower preoperative lymphocyte percentage (p = 0.022), higher NLR (p = 0.004) and PLR (p = 0.001), an increased PRBC transfusions (p = 0.001), and diabetes mellitus (p = 0.033). Multivariate analysis with automatic linear modeling showed that only the preoperative PLR was a powerful predictor for the delay of feeding resumption (p < 0.01).ConclusionThe lymphocyte percentage, PLR, and NLR are found to be associated with prolonged hospitalization and ICU stay and delayed oral feeding following THE for Siewert types II and III AEG. We hope by this series, to have set, at least one preliminary cornerstone, in the creation of a prognostic model, capable of assessing the need for an intraoperative jejunostomy placement, in patients undergoing esophagectomy for distal esophageal carcinoma.

Long term combined fertilization and soil aggregate size on the denitrification and community of denitrifiers

Balanced fertilization strategies that combine the use of organic and inorganic fertilizers have been adopted to increase soil fertility and crop yield and to safeguard environment. However, knowledge on the influence of these fertilization strategies on microbe-mediated nitrogen cycling in the agricultural ecosystems remains limited. We analyzed the denitrifiers with nitrite reductase genes (nirK and nirS) in three soil aggregate classes (large macroaggregates, >2 mm; small macroaggregates, 0.25–2 mm; and microaggregates, <0.25 mm) in response to long-term application of inorganic fertilizers (NPK), NPK plus straw (NPKS), and NPK plus manure (NPKM) in the fluvo-aquic soil of North China Plain. Both fertilization and aggregate size had significant effects on the nirK and nirS gene abundances and soil potential denitrification activity (PDA). PDA in the combined fertilizer-treated soils (NPKS and NPKM treatments) was two times more than that in inorganic NPK-treated soil. PDA dramatically increased in small macroaggregates and microaggregates of NPKS treatment. The nirS gene abundance was significantly higher than that of the nirK gene abundance, and NPKS and NPKM treatments increased the nirS to nirK ratio in bulk soils and in microaggregates compared with NPK treatment and control. Correlation analysis and automatic linear modeling revealed a primary correlation between PDA and abundance of nirS-type denitrifiers. Soil properties, such as SOC, pH and TN, significantly affected the community structure of nirK- and nirS-type denitrifiers while only aggregate size significantly influenced the composition of nirK-type denitrifiers. Taken together, our results suggest that long-term combined fertilization increased the denitrification potential in small macroaggregate and microaggregate classes, and the increased gene abundance and the shift community structures of nirS- and nirK-type denitrifiers were potentially responsible for an increase in the promoted denitrification.

Individual predictors of tactical planning performance in junior military officers

ABSTRACTThis exploratory study examined the influence of personal and psycholinguistic factors on tactical planning performance in a convenience sample of Australian Army personnel. Sixty male junior officers undertaking a tactical training course completed self-report measures of decision-making style, problem-solving appraisal, personality, and tactical planning expertise. A measure of general cognitive ability was sourced from participants’ military records. During the course, officers completed several tactical planning exercises that were assessed by course instructors, and an overall measure of tactical planning performance was derived. Psycholinguistic data was derived from participants’ thought processes elicited during an analogous planning exercise using a written think-aloud protocol. Automatic linear modeling analyses identified the combination of greater expertise and intuitive decision-making style, less spontaneous decision-making style and problem-solving confidence, younger age, and greater use of words relating to risk and past-focused concepts in their thought records, as significant predictors of better tactical planning, accounting for 55% of the variance in performance. Additionally, general cognitive ability, conscientiousness, and rational decision-making style shared moderate positive bivariate correlations with planning performance. This research has identified individual factors that contribute to tactical mission planning, along with future research avenues, which may inform the training of military commanders at the tactical level.

Identifying Fraction Measures as Screeners of Mathematics Risk Status

This study investigated the accuracy of three fraction measures (i.e., fraction number line estimation accuracy, general fraction concepts, and fraction arithmetic) for screening fourth graders who might be at risk for mathematics difficulties. Receiver operating characteristic (ROC) curve analyses assessed diagnostic accuracy of the fraction measures for predicting which students would not meet state standards on the state mathematics test in fourth grade (n = 411), fifth grade (n = 362), and sixth grade (n = 304). A combined measure consisting primarily of fraction number line estimation items and general fraction concept items was the most accurate screener of risk status in fourth, fifth, and sixth grades (area under the curve [AUC] = .84, .81, and .85, respectively). To maximize efficiency for classroom use, the length of the combined screener was reduced using best subset automatic linear modeling. The study highlights the importance of fraction knowledge for predicting mathematics achievement more generally and validates an effective and practical screening tool for the intermediate grades.

Forecasting banking sectors in Indian stock markets using machine intelligence

The study analyze stock index closing from myriad set of technical and fundamental analysis variables extracted from real market data to assist forecast of market closing. For this, major service sector indices of Bombay stock exchange (BSE) and National stock exchange (NSE) with historical data we re taken from banking industry. The predictive model performance of index closing using statistical procedures like automatic linear modeling, time-series based econometric forecasting, vector auto regression as with artificial neural network based models were simulated and analyzed. The results indicate that BSE had higher forecast accuracy using autoregressive models and market volatility factor had major influence. Whereas, NSE was impacted by quarterly performance that can be modeled using neural networks. The empirical results were contrasted with latest state-of-art research theories to provide agenda and future research challenges of market forecast systems.

Nitrous oxide emission, global warming potential, and denitrifier abundances as affected by long-term fertilization on Mollisols of Northeastern China

ABSTRACTA long-term field experiment was performed to assess the effects of fertilization regimes on greenhouse gas emissions, soil properties, soil denitrifies, and maize (Zea mays) grain yield on Mollisols of Northeastern China. Chemical nitrogen (N), phosphorus (P), and potassium (K) fertilizers plus pig manure (MNPK) treatment significantly increased soil N2O emissions by 29.9–226.4% and global warming potential (GWP) by 29.8–230.7% compared to unfertilized control (CK), chemical N fertilizer only (N), chemical N, P, and K fertilizers (NPK) and chemical N, P, and K fertilizers plus corn straw (SNPK) treatments. However, the MNPK treatment yielded similar greenhouse gas intensity (GHGI) as compared with other treatments, mainly due to higher maize grain yield. There were also higher gene copy numbers of nirK, nirS, and nosZ in topsoil (0–20 cm depth) under MNPK treatment. Automatic linear modeling analysis indicated that main factors influencing soil N2O emissions were soil organic carbon (SOC), NO3− content, and nirK gene abundance. Although the application of chemical fertilizers plus organic manure increases N2O emissions due to higher N and C availability and nirK gene activity in the soil, this is still a promising fertilizer management due to its notable enhancement of maize grain yield and SOC content.

Older Adults' Live Demonstration of Electronic Personal Health Record Use: Factors Mediating Initial Proficiency.

Adoption of electronic personal health records by older adults offers multiple advantages to healthcare and is being encouraged by federal agencies and health associations. However, obstacles have limited older adults' rates of adoption to approximately 10%. This study examined the initial proficiency of older adults at entering 21 standard health elements into an electronic personal health record. Entry completeness, accuracy, elapsed time, and help requests were measured. A combination of standard technology adoption model and older adult characteristics accounted for 52% of variability in proficiency at entering electronic personal health record data. Automatic linear modeling identified three variables as primarily related to proficiency with electronic personal health record use: age, computer competency, and mental status. Interventions to increase electronic personal health record adoption and proficiency of use will require the consideration of variables specific to older adults, and may best focus on younger seniors with good mental status and computer competency. Efforts for older seniors with decreased mental status might better center on delegation to a primary caregiver.

Difficulty scoring system in laparoscopic distal pancreatectomy.

Several factors affect the level of difficulty of laparoscopic distal pancreatectomy (LDP). The purpose of this study was to develop a difficulty scoring (DS) system to quantify the degree of difficulty in LDP. We collected clinical data for 80 patients who underwent LDP. A 10-level difficulty index was developed and subcategorized into a three-level difficulty index; 1-3 as low, 4-6 as intermediate, and 7-10 as high index. The automatic linear modeling (LINEAR) statistical tool was used to identify factors that significantly increase level of difficulty in LDP. The operator's 10-level DS concordance between the 10-level DS by the reviewers, LINEAR index DS, and clinical index DS systems were analyzed, and the weighted Cohen's kappa statistic were at 0.869, 0.729, and 0.648, respectively, showing good to excellent inter-rater agreement. We identified five factors significantly affecting level of difficulty in LDP; type of operation, resection line, proximity of tumor to major vessel, tumor extension to peripancreatic tissue, and left-sided portal hypertension/splenomegaly. This novel DS for LDP adequately quantified the degree of difficulty, and can be useful for selecting patients for LDP, in conjunction with fitness for surgery and prognosis.

Executive and social-cognitive determinants of environmental dependency syndrome in behavioral frontotemporal dementia.

Environmental dependency syndrome (EDS), including utilization (UB) and imitation (IB) behaviors, is often reported in behavioral variant frontotemporal dementia (bvFTD). These behaviors are commonly attributed to executive dysfunction. However, inconsistent associations between EDS and poor executive performance has led to an alternative "social hypothesis," instead implicating patients' misinterpretation of the examiner's intention. We investigated the possible explanatory cognitive mechanisms of EDS in bvFTD by relating UB and IB to performance on tests of executive functioning and theory of mind (ToM). This study analyzed retrospective data of 32 bvFTD patients. Data included scores of UB and IB, various executive measures, and ToM assessment using the faux pas test, from which we extracted a mental attribution score. Of the patients, 15.6% and 40.6% exhibited UB and IB, respectively. We conducted an automatic linear modeling analysis with executive and mental attribution measures as predictor variables, and UB and IB sequentially considered as target variables. ToM mental attribution score, visual abstraction and flexibility measures from the Wisconsin Card Sorting Test, and motor sequence performance significantly (corrected ps < .05) predicted IB. No executive or ToM measures significantly predicted UB. These findings reveal a complex interaction between executive dysfunction and mental attribution deficits influencing the prevalence of EDS in bvFTD. Further investigation is required to improve our understanding of the mechanisms underlying these behaviors. (PsycINFO Database Record

Predicting heat stress index in Sasso hens using automatic linear modeling and artificial neural network.

There is an increasing use of robust analytical algorithms in the prediction of heat stress. The present investigation therefore, was carried out to forecast heat stress index (HSI) in Sasso laying hens. One hundred and sixty seven records on the thermo-physiological parameters of the birds were utilized. They were reared on deep litter and battery cage systems. Data were collected when the birds were 42- and 52-week of age. The independent variables fitted were housing system, age of birds, rectal temperature (RT), pulse rate (PR), and respiratory rate (RR). The response variable was HSI. Data were analyzed using automatic linear modeling (ALM) and artificial neural network (ANN) procedures. The ALM model building method involved Forward Stepwise using the F Statistic criterion. As regards ANN, multilayer perceptron (MLP) with back-propagation network was used. The ANN network was trained with 90% of the data set while 10% were dedicated to testing for model validation. RR and PR were the two parameters of utmost importance in the prediction of HSI. However, the fractional importance of RR was higher than that of PR in both ALM (0.947 versus 0.053) and ANN (0.677 versus 0.274) models. The two models also predicted HSI effectively with high degree of accuracy [r = 0.980, R2 = 0.961, adjusted R2 = 0.961, and RMSE = 0.05168 (ALM); r = 0.983, R2 = 0.966; adjusted R2 = 0.966, and RMSE = 0.04806 (ANN)]. The present information may be exploited in the development of a heat stress chart based largely on RR. This may aid detection of thermal discomfort in a poultry house under tropical and subtropical conditions.

Nonchemotherapy drug-induced agranulocytosis in children detected by a prospective pharmacovigilance program

ABSTRACTObjectives: A prospective evaluation of nonchemotherapy drug-induced agranulocytosis (DIA) cases, which are infrequent in the pediatric population. We characterize agranulocytosis cases and assess lab test differences between drug- and nondrug-induced agranulocytosis. Methods: Through our Prospective Pharmacovigilance Program from Laboratory Signals at Hospital we detected pediatric agranulocytosis cases from July 2007 to December 2010. This program estimates the incidence, drug causality, clinical features, outcomes of DIA pediatric cases, and assesses laboratory differences with respect to non-DIA. Results: We detected 662 agranulocytosis in 308 pediatric patients, of which 14 were caused by nonchemotherapy drugs. The incidence rate of DIA for 10,000 pediatric patients was 3.92 (Poisson 95% confidence interval 1.09–8.77); 78.6% of DIA cases occurred in patients younger than 3 years. The final outcome was recovery without sequela in all cases. The pharmacologic group most frequently implicated was antimicrobial drugs (11 drugs), 7 of which were beta-lactams. The drugs most frequently suspected were cefotaxime and vancomycin (3 cases each). We found 3 drugs (cloperastine, codeine, and enoxaparin) not previously described to induce DIA. Automatic linear modeling (n = 56, R2 = 45.2%) showed a significant inverse association with platelets (R2 = 17.5%), hemoglobin, and alanine transaminase, and a direct association with red cell distribution (R2 = 16.2%). A generalized linear model (Type III, n = 1188; DIA, n = 86; likelihood ratio chi-squared = 156.16) retained eosinophils (p <.001), platelets (p <.001), total serum proteins (p <.001), and hemoglobin (p =.039). Conclusions: We found a higher incidence of DIA in children than previously described. Our findings also suggest an immune-mediated destruction or myeloid toxicity, possibly facilitated by an increase in drug exposure.

Increased homocysteine plasma level is associated with shortened prothrombin time in HIV-infected patients

Objective: To find factors associated with increased homocysteine plasma level in HIV-infected patients.Methods: Cross-sectional study, carried out as a supplementary task to the standard care of HIV-infected patients. The possible association of increased homocysteine plasma level with blood analyses results was assessed with a multiple linear regression analysis, using the automatic linear modeling available in SPSS version 22.Results: A total of 145 patients were included. Creatinine was higher than normal in 7 patients (5%), prothrombin time was shortened in 36 patients (25%), and a monoclonal gammopathy was detected in 2 patients (1%). In the regression analysis, an association was found between high homocysteine plasma level and the following variables: low prothrombin time (β coefficient −0.286, confidence interval −1.1854 to −0.754, p < 0.001), high creatinine (coefficient 9.926, confidence interval 6.351–15.246, p < 0.001), low folic acid (coefficient −0.331, confidence interval −0–483 to −0.187, p < 0.001), and low vitamin B12 (coefficient −0.007, confidence interval −0.01 to −0.001, p = 0.005).Conclusion: An association was found between increased homocysteine plasma level and shortened prothrombin time.

Perfusion Neuroimaging Abnormalities Alone Distinguish National Football League Players from a Healthy Population.

Background: National Football League (NFL) players are exposed to multiple head collisions during their careers. Increasing awareness of the adverse long-term effects of repetitive head trauma has raised substantial concern among players, medical professionals, and the general public.Objective: To determine whether low perfusion in specific brain regions on neuroimaging can accurately separate professional football players from healthy controls.Method: A cohort of retired and current NFL players (n = 161) were recruited in a longitudinal study starting in 2009 with ongoing interval follow up. A healthy control group (n = 124) was separately recruited for comparison. Assessments included medical examinations, neuropsychological tests, and perfusion neuroimaging with single photon emission computed tomography (SPECT). Perfusion estimates of each scan were quantified using a standard atlas. We hypothesized that hypoperfusion particularly in the orbital frontal, anterior cingulate, anterior temporal, hippocampal, amygdala, insular, caudate, superior/mid occipital, and cerebellar sub-regions alone would reliably separate controls from NFL players. Cerebral perfusion differences were calculated using a one-way ANOVA and diagnostic separation was determined with discriminant and automatic linear regression predictive models.Results: NFL players showed lower cerebral perfusion on average (p < 0.01) in 36 brain regions. The discriminant analysis subsequently distinguished NFL players from controls with 90% sensitivity, 86% specificity, and 94% accuracy (95% CI 95-99). Automatic linear modeling achieved similar results. Inclusion of age and clinical co-morbidities did not improve diagnostic classification.Conclusion: Specific brain regions commonly damaged in traumatic brain injury show abnormally low perfusion on SPECT in professional NFL players. These same regions alone can distinguish this group from healthy subjects with high diagnostic accuracy. This study carries implications for the neurological safety of NFL players.

云南省彝语地名与自然社会因子的关系分析

为探究云南省彝语地名与自然社会因子的关系，采用相关分析、自动线性建模、空间回归等方法进行不同侧面的分析。分析结果如下：1) 从相关分析与回归分析结果来看，各项自然社会因子重要性从大到小依次排列为：彝语地名与铁路距离、高程、彝语地名与道路距离、彝语地名与农用地距离、彝语地名与主要城市距离、彝语地名与水系距离、坡度。2) 从空间回归结果看，地区生产总值、人均GDP、年均温、年降水量、人均受教育年限、文盲率、成人识字率、彝族人口数量与彝语地名数量成负相关。第一产业增加值、第二产业增加值、第三产业增加值、医疗机构床位数、总人口与彝语地名数量成正相关。医疗机构床位数、年降水量通过显著性检验，表明彝语地名数量多的县市区医疗机构床位数相对也多，医疗条件相对较好；彝语地名数量多的县市区降水量相对较少。 In order to explore relationship between natural and social factors and placenames named in Yi nationality language in Yunnan province, the paper used correlation analysis, automatic linear modeling and spatial regression analysis methods to analyze it, then these conclusions could be drawn as follows. 1) From the correlation analysis and regression analysis results, the importance of natural and social factors were arranged in descending order: the distance between placenames named in Yi nationality language and railway, elevation, the distance between placenames and road, the distance between placenames and agricultural land, the distance between placenames and main cities, the distance between placenames and water system, slope. 2) From the spatial regression analysis results, GDP, real GDP per capita, annual mean temperature, annual precipita-tion, education years’ per-capita, illiteracy rate, adult literacy rate and size of Yi population were negatively correlated to the number of placenames named in YI nationality language. The value- added of the primary industry, value-added of secondary industry, value-added of the tertiary in-dustry, number of beds in medical institutions and total population were positively correlated to the number of placenames named in Yi nationality language. And the number of beds in medical institutions and annual precipitation passed the significance test, this indicated the counties and cities that had more placenames named in Yi nationality language had more beds in medical in-stitutions, they had relatively good medical conditions, however, their annual precipitation was relatively little.

Neural Network Prediction of ICU Length of Stay Following Cardiac Surgery Based on Pre-Incision Variables.

BackgroundAdvanced predictive analytical techniques are being increasingly applied to clinical risk assessment. This study compared a neural network model to several other models in predicting the length of stay (LOS) in the cardiac surgical intensive care unit (ICU) based on pre-incision patient characteristics.MethodsThirty six variables collected from 185 cardiac surgical patients were analyzed for contribution to ICU LOS. The Automatic Linear Modeling (ALM) module of IBM-SPSS software identified 8 factors with statistically significant associations with ICU LOS; these factors were also analyzed with the Artificial Neural Network (ANN) module of the same software. The weighted contributions of each factor (“trained” data) were then applied to data for a “new” patient to predict ICU LOS for that individual.ResultsFactors identified in the ALM model were: use of an intra-aortic balloon pump; O2 delivery index; age; use of positive cardiac inotropic agents; hematocrit; serum creatinine ≥ 1.3 mg/deciliter; gender; arterial pCO2. The r2 value for ALM prediction of ICU LOS in the initial (training) model was 0.356, p <0.0001. Cross validation in prediction of a “new” patient yielded r2 = 0.200, p <0.0001. The same 8 factors analyzed with ANN yielded a training prediction r2 of 0.535 (p <0.0001) and a cross validation prediction r2 of 0.410, p <0.0001. Two additional predictive algorithms were studied, but they had lower prediction accuracies. Our validated neural network model identified the upper quartile of ICU LOS with an odds ratio of 9.8(p <0.0001).ConclusionsANN demonstrated a 2-fold greater accuracy than ALM in prediction of observed ICU LOS. This greater accuracy would be presumed to result from the capacity of ANN to capture nonlinear effects and higher order interactions. Predictive modeling may be of value in early anticipation of risks of post-operative morbidity and utilization of ICU facilities.

Examination of the Relationship between General Self-efficacy Beliefs, Emotional Intelligence Levels and Emotional Self-efficacy Levels of Students in School of Physical Education and Sport

The aim of the present research paper is to examine the relationship between general self-efficacy (GSE) beliefs, emotional intelligence (EI) levels and emotional self-efficacy (ESE) levels of students in school of physical education and sport. 256 students in school of physical education and sport at Canakkale Onsekiz Mart University participated in the study. A model was hypothesized and model fit indices were examined in AMOS. Significant differences were noted between departments of physical education and sports teacher and coach education in terms of positive regulation. Positive correlations were found between ESE, EI and GSE. The hypothesized model could not be accepted; however after applying automatic linear modeling, it was observed that EI was the most important predictor of ESE. Consequently, linear modeling analyses have shown that EI and GSE are important predictors of ESE.