Six-variable Model Research Articles

Derivation and validation of a risk score to predict acute kidney injury in critically ill cirrhotic patients.

Acute kidney injury (AKI) is a common complication in critically ill cirrhotic patients with substantial mortality. Given AKI can be prevented via early detection, it is urgent to develop an easy model to identify high-risk patients. A total of 1149 decompensated cirrhotic (DC) patients from the eICU Collaborative Research Database were enrolled for model development and internal validation. The variables used for analysis mainly included laboratory tests. We first built an ensemble model (random forest, gradient boosting machine, K-nearest neighbor, and artificial neural network) named DC-AKI using machine learning methods. Based on the Akaike Information Criterion (AIC), we then constructed a risk score, which was externally validated in 789 DC patients from the Medical Information Mart for Intensive Care database. AKI developed in 212 (26%) of 804 patients in the derivation cohort and 355 (45%) of 789 patients in the external validation cohort. DC-AKI identified the 8 variables most strongly associated with the outcome: serum creatinine (sCr), total bilirubin, magnesium, shock index (SI), prothrombin time, mean corpuscular hemoglobin, lymphocytes and arterial oxygen saturation (SaO2 ). Based on the smallest AIC, a 6-variable model was eventually used to construct the scoring system (sCr, total bilirubin, magnesium, SI, lymphocytes and SaO2 ). The scoring system showed good discrimination with the AUROC of 0.805 and 0.772 in two validation cohorts. The scoring system using routine laboratory data was able to predict the development of AKI in critically ill cirrhotic patients. The utility of this score in clinical care requires further research. This article is protected by copyright. All rights reserved.

Clear sky color modeling based on BP neural network

Daylight is a valuable resource that provides energy and luminance, as it contains changing spectrum distributions, which also carry color information and have an important impact on human vision and perception. Several sky models have been proposed for use in daylighting simulation calculations to quantify the effect of daylight. However, these models do not fully consider the color, or are modeled using the correlation of the color temperature and luminance fitting, which has color deviation, so it is difficult to realize the chromaticity-related research of daylight. Therefore, the clear sky spectral data in the Beijing area was measured and used to calculate the color coordinates of sky patches. Corresponding it with the sun position, sky patch position, and radiation data to model the sky color distribution based on backpropagation neural network method. Four-, five-, and six-variable models were established, where the five-variable model exhibited the smallest error: NMBEx = 0.52%, CVRMSEx = 3.54%, NMBEy = 0.67%, and CVRMSEy = 3.55%. However, the four- and six-variable errors were also within the specification and were easier to use. The proposed models can achieve the real-time prediction of the sky color distribution, provide a theoretical basis for daylighting design in future buildings, and guide the development of new spectrally-selective material applications and intelligent lighting control systems.

Distance education students’ acceptance of online learning systems, attitudes towards online learning and their self-directed learning skills

About four million people living in Turkey or abroad with a Turkish passport enrolled in distance education schools to have the opportunity to complete their compulsory formal education in the spring of 2018-19. The aim of this descriptive study was to determine distance education students’ acceptance of online learning systems, their attitudes towards online learning, and their self-directed learning skills and the relationship between them. The study group consisted of 1233 students in the spring term of 2018-2019. Some of the main results are as follows: A six-variable model (perceived ease of use, perceived usefulness, general acceptance, individual awareness, usefulness, application effectiveness) played a significantly explanatory role in dependent variables (subscales of self-directed learning skills). Perceived usefulness and usefulness positively predicted motivation, suggesting that the higher the perceived usefulness, the higher the motivation. General acceptance and application effectiveness negatively predicted self-monitoring. Female participants were more motivated by online learning environments than males.

Pediatric severe traumatic brain injury mortality prediction determined with machine learning-based modeling

IntroductionSevere traumatic brain injury (sTBI) is a leading cause of mortality in children. As clinical prognostication is important in guiding optimal care and decision making, our goal was to create a highly discriminative sTBI outcome prediction model for mortality. MethodsMachine learning and advanced analytics were applied to the patient admission variables obtained from a comprehensive pediatric sTBI database. Demographic and clinical data, head CT imaging abnormalities and blood biochemical data from 196 children and adolescents admitted to a tertiary pediatric intensive care unit (PICU) with sTBI were integrated using feature ranking by way of a forest of randomized decision trees, and a model was generated from a reduced number of admission variables with maximal ability to discriminate outcome. ResultsIn total, 36 admission variables were analyzed using feature ranking with variable weighting to determine their predictive importance for mortality following sTBI. Reduction analysis utilizing Borata feature selection resulted in a parsimonious six-variable model with a mortality classification accuracy of 82%. The final admission variables that predicted mortality were: partial thromboplastin time (22%); motor Glasgow Coma Scale (21%); serum glucose (16%); fixed pupil(s) (16%); platelet count (13%) and creatinine (12%). Using only these six admission variables, a t-distributed stochastic nearest neighbor embedding algorithm plot demonstrated visual separation of sTBI patients that lived or died, with high mortality predictive ability of this model on the validation dataset (AUC = 0.90) which was confirmed with a conventional area-under-the-curve statistical approach on the total dataset (AUC = 0.91; P < 0.001). ConclusionsMachine learning-based modeling identified the most clinically important prognostic factors resulting in a pragmatic, high performing prognostic tool for pediatric sTBI with excellent discriminative ability to predict mortality risk with 82% classification accuracy (AUC = 0.90). After external multicenter validation, our prognostic model might help to guide treatment decisions, aggressiveness of therapy and prepare family members and caregivers for timely end-of-life discussions and decision making. Level of evidenceIII; Prognostic.

Predicting Quality and Yield of Growing Alfalfa from a UAV

HighlightsThe canopy height distributions from photogrammetry were mostly Gaussian distributions.The mean canopy height and standard deviation from photogrammetry can be used to predict yield and nutritive values.Including average field maturity and pest pressures improves the models to R2 values of around 0.8.The best predictive model types were generally Gaussian random processes.Abstract. Alfalfa producers would be able to manage their crop production practices better if they knew the distribution of yield and nutritive values of the alfalfa growing throughout their fields. Unmanned aerial vehicles (UAVs) equipped with cameras and photogrammetry techniques provide methods to quickly capture the plant canopy structure at field scale. The goal of this study was to determine how to use the point clouds produced by the photogrammetry process to estimate the yield and nutritive value of alfalfa throughout its growth cycle. During the 2017 growing season, weekly measurements were taken of 1 m2 quadrats (~20 per week, 325 total) in a field of alfalfa managed for forage production. Measurements in each quadrat included manual measurements of maximum and average height, weed presence, disease damage, insect damage, maturity level, stand plant density, and many images of the quadrat from a UAV. After processing to remove outliers, the canopy heights from the photogrammetry point clouds were largely Gaussian distributions. Models were developed using supervised machine learning to estimate yield and nutritive values, including acid detergent fiber (ADF), neutral detergent fiber (NDF), and crude protein (CP), with different numbers of predictor (input) variables. Simple models with two predictor variables were only based on the mean and standard deviation of the heights of the photogrammetry point cloud. The models with three predictor variables added average field maturity level. Finally, the models with six predictor variables added weed presence, insect damage, and disease damage. A linear regression with all interaction terms was found to be the best type of model for predicting yield with six variables. For all other outputs and numbers of predictor variables, a Gaussian random process (GRP) model was best. The models improved with additional predictor variables, so the six-variable models were best able to predict yield and nutritive value. The R2 values for the six-variable models for predicting yield, ADF, NDF, and CP were 0.81, 0.81, 0.78, and 0.79, respectively. Keywords: Alfalfa, Machine learning, Nutritive value, Photogrammetry, Unmanned aerial vehicle, Yield.

Using best subset regression to identify clinical characteristics and biomarkers associated with sepsis-associated acute kidney injury.

Sepsis-associated acute kidney injury (AKI) is a complex clinical disorder associated with inflammation, endothelial dysfunction, and dysregulated coagulation. With standard regression methods, collinearity among biomarkers may lead to the exclusion of important biological pathways in a single final model. Best subset regression is an analytic technique that identifies statistically equivalent models, allowing for more robust evaluation of correlated variables. Our objective was to identify common clinical characteristics and biomarkers associated with sepsis-associated AKI. We enrolled 453 septic adults within 24 h of intensive care unit admission. Using best subset regression, we evaluated for associations using a range of models consisting of 1-38 predictors (composed of clinical risk factors and plasma and urine biomarkers) with AKI as the outcome [defined as a serum creatinine (SCr) increase of ≥0.3 mg/dL within 48 h or ≥1.5× baseline SCr within 7 days]. Two hundred ninety-seven patients had AKI. Five-variable models were found to be of optimal complexity, as the best subset of five- and six-variable models were statistically equivalent. Within the subset of five-variable models, 46 permutations of predictors were noted to be statistically equivalent. The most common predictors in this subset included diabetes, baseline SCr, angiopoetin-2, IL-8, soluble tumor necrosis factor receptor-1, and urine neutrophil gelatinase-associated lipocalin. The models had a c-statistic of ∼0.70 (95% confidence interval: 0.65-0.75). In conclusion, using best subset regression, we identified common clinical characteristics and biomarkers associated with sepsis-associated AKI. These variables may be especially relevant in the pathogenesis of sepsis-associated AKI.

Why the Increasing Use of Complex Causal Models Is a Problem: On the Danger Sophisticated Theoretical Narratives Pose to Truth

Causal models in organizational research are complex. As use of complex models increases, the joint probability a published model is true decreases. Across The Academy of Management Journal ( AMJ), Organizational Behavior and Human Decision Processes ( OBHDP), and Administrative Science Quarterly ( ASQ) from 2016 to 2018, it was most common to see six variables in a causal model. Even with a generous 80% independent probability of each correlation being properly theorized, the joint probability of a six-variable model is about 3.5%. Further, causal models often involve a causal chain, rendering the model even more improbable. Consequently, much of the knowledge generated in top journals is likely false. We explain that peer review demands for sophisticated theoretical narratives may pressure researchers to produce models that are embarrassingly unlikely. Traditionally, researchers argue that a low probability model is overcome by prior theory. Using an ethnostatistical Bayesian analysis, we found that given a generous prior likelihood ratio of 20, the posterior likelihood ratio is less than 1. Finally, we add “not reporting belief in a complex model” to the domain of questionable research practices and discuss auxiliary assumptions, the unstated assumptions that contextualize a theory. To ease reporting on belief in a complex model please see the following calculator: https://practiceoftheory.weebly.com/a-causal-models-probability-of-being-true.html .

Monetary Policy Transmission Mechanisms in the Central and Eastern European Countries

Objective: The aim of this study is to examine monetary policy transmission mechanisms in four Central and Eastern European countries (the Czech Republic, Hungary, Poland and Romania), in the presence of fiscal and exchange rate effects. Research Design & Methods: We implement a structural vector autoregression (SVAR) approach for modelling the interdependencies between monetary and fiscal policies, output gap and consumer price inflation (CPI). In our six-variable model, which includes the budget balance, the output gap, CPI, the central bank reference rate, the lending rate and the real exchange rate (RER), short-run restrictions on the contemporaneous structural parameters imply that the budget balance responds to changes in the output gap and lending rate, while the central bank reference rate is a function of output and inflationary shocks. Findings: The results of our research show that the effects of an increase in the central bank’s short-run interest rate on inflation, output gap and the RER are quite heterogeneous across the CEE countries. As the monetary policy response to inflation seems to be significant and rather uniform across countries, though with a different time pattern, there is no evidence of its reaction to the output gap (except for Romania in the long run). Among other results, budget surplus has a strong anti-inflationary impact in all countries but at the expense of a short-lived output slowdown (except for Hungary). The RER undervaluation is likely to stimulate output (Romania) or depress it (Poland), with a neutral stance in the two other countries. As expected, an increase in the lending rate is followed by a fall in output on impact, while there is no significant effect on inflation. Implications/Recommendations: Our study argues in favour of a much stronger response of the central bank reference rate to the output gap in the CEE countries. As suggested by the experience of Poland, an immediate response of the central bank to inflation could explain the lack of the price puzzle when an increase in the reference rate is associated with a sustained increase in consumer prices. An anti-inflationary monetary policy stance should be strengthened by fiscal tightening, while in a recession a higher budget deficit is likely to boost output and prevent a deflationary spiral. Contribution: The article presents the application of the extended IS-MP-IA model to the modelling of monetary policies by the central banks of the CEE countries that practice a floating exchange rate regime.

Predicting CBR Values of Black Cotton Soil Stabilized with Cement and Waste Glass Admixture Using Regression Model

In highway constructions, sub-grade and sub-base soil stabilization has been used as one of the prime and major process for many years in order to improve the engineering properties of soil. The strength of theses layers is indicated by their California bearing ratio (CBR) value which is quite expensive and time consuming. In order to overcome this situation, this study presents a methodology for predicting soaked California Bearing Ration (CBR) value of Black Cotton Soil Stabilized with Cement and Waste Glass Admixture using Multiple Regression Analysis (MRA). Experimental test results such Atterberg limit (Liquid limit (LL), Plastic limit (PL) and Plasticity index (PI)), Compaction characteristics of two compactive efforts namely standard proctor (SP) and modified proctor (MP) (maximum dry density (MDD) and optimum moisture content (OMC)), CBR, Waste glass (WG) content and Cement content (Cm), obtained from a laboratory at Abubakar Tafawa Balewa University Bauchi, Nigeria, have been employed in developing multiple regression models. California Bearing Ration was taken as the dependent variables while Liquid limit, Plastic limit, maximum dry density, optimum moisture content, waste glass content and Cement content were taken as independent variables. The regression analysis calculated the error mean square (MSE) for each possible model, and models with large MSE were not selected for the best regression equations. The best models have a minimum value of MSE occurring for the six-variable model (Cm, WG, LL, PL, OMCsp, MDDsp) and (Cm, WG, PL, LL, OMCmp, MDDmp) with a corresponding higher value of coefficient of multiple determination R2 = 0.98 and 0.94. The performance evaluation of the fitted regression models indicates a strong correlation (R2 = 0.89 - 0.98) between the mentioned variables, and the model equations developed from this work provided a very good prediction of the response, as the equations can be employed for making estimates of soaked CBR of other black cotton soils having similar geotechnical properties.

Sex determination of a Tunisian population by CT scan analysis of the skull.

It is widely accepted that the estimation of biological attributes in the human skeleton is more accurate when population-specific standards are applied. With the shortage of such data for contemporary North African populations, it is duly required to establish population-specific standards. We present here the first craniometric standards for sex determination of a contemporary Tunisian population. The aim of this study was to analyze the correlation between sex and metric parameters of the skull in this population using CT scan analysis and to generate proper reliable standards for sex determination of a complete or fragmented skull. The study sample comprised cranial multislice computed tomography scans of 510 individuals equally distributed by sex. ASIRTM software in a General ElectricTM workstation was used to position 37 landmarks along the volume-rendered images and the multiplanar slices, defining 27 inter-landmark distances. Frontal and parietal bone thickness was also measured for each case. The data were analyzed using basic descriptive statistics and logistic regression with cross-validation of classification results. All of the measurements were sexually dimorphic with male values being higher than female values. A nine-variable model achieved the maximum classification accuracy of 90% with -2.9% sex bias and a six-variable model yielded 85.9% sexing accuracy with -0.97% sex bias. We conclude that the skull is highly dimorphic and represents a reliable bone for sex determination in contemporary Tunisian individuals.

Predicting the 6-month risk of severe hypoglycemia among adults with diabetes: Development and external validation of a prediction model

AimsTo develop and externally validate a prediction model for the 6-month risk of a severe hypoglycemic event among individuals with pharmacologically treated diabetes. MethodsThe development cohort consisted of 31,674 Kaiser Permanente Colorado members with pharmacologically treated diabetes (2007–2015). The validation cohorts consisted of 38,764 Kaiser Permanente Northwest members and 12,035 HealthPartners members. Variables were chosen that would be available in electronic health records. We developed 16-variable and 6-variable models, using a Cox counting model process that allows for the inclusion of multiple 6-month observation periods per person. ResultsAcross the three cohorts, there were 850,992 6-month observation periods, and 10,448 periods with at least one severe hypoglycemic event. The six-variable model contained age, diabetes type, HgbA1c, eGFR, history of a hypoglycemic event in the prior year, and insulin use. Both prediction models performed well, with good calibration and c-statistics of 0.84 and 0.81 for the 16-variable and 6-variable models, respectively. In the external validation cohorts, the c-statistics were 0.80–0.84. ConclusionsWe developed and validated two prediction models for predicting the 6-month risk of hypoglycemia. The 16-variable model had slightly better performance than the 6-variable model, but in some practice settings, use of the simpler model may be preferred.

Did Quantitative Easing Only Inflate Stock Prices? Macroeconomic Evidence from the US and UK

This paper considers the impact of US and UK Quantitative Easing (QE) on their respectiveeconomies with a particular focus on the stock market, production and price levels. We conduct anempirical quantitative exercise based on a novel six-variable VAR model, which combines macroeconomic and forward-looking financial variables and uses a 'pure' measure of QE. The results suggest a positive response of equity prices, and a 'V' shaped reaction of volatility and the bid-ask spread to the monetary stimulus. Output and inflation, in contrast with some previous studies, show an insignificant impact providing evidence of the limitations of the central bank's programmes. We attribute the variation to this difference in our modelling approach, which includes stock market variables, and we conclude that its presence is of critical importance in the assessment of unconventional monetary policy. Economically, we argue that the reason for the negligible economic stimulus of QE is that the money injected funded financial asset price growth more than real projects.

Complex bifurcations in the oscillatory reaction model

The mixing of different types of bifurcations, i.e. supercritical Andronov–Hopf (SAH), double loop (DL) and saddle-loop (SL) bifurcations in the vicinity of their total annihilation, is examined on the highly nonlinear six-variable model for the Bray–Liebhafsky (BL) oscillatory reaction under continuously well-stirred tank reactor (CSTR) conditions. For this kind of the reaction system where the law of mass conservation is additional constraint that must be satisfied and where because of that, some simple bifurcations cannot be formed independently to the others, the considered transformations of the bifurcations are particularly important. That is why as the control parameters for bifurcation analysis, the specific flow rate (j0), as well as the inflow hydrogen peroxide concentration (h=[H2O2]in), were used. The complex bifurcations obtained from numerical simulations are compared with some experimental results. It was shown that these complex bifurcations cannot be easily recognized in experimental investigations without knowing their evolution.

Predicting Non-Adherence with Outpatient Colonoscopy Using a Novel Electronic Tool that Measures Prior Non-Adherence.

Accurately predicting the risk of no-show for a scheduled colonoscopy can help target interventions to improve compliance with colonoscopy, and thereby reduce the disease burden of colorectal cancer and enhance the utilization of resources within endoscopy units. We aimed to utilize information available in an electronic medical record (EMR) and endoscopy scheduling system to create a predictive model for no-show risk, and to simultaneously evaluate the role for natural language processing (NLP) in developing such a model. This was a retrospective observational study using discovery and validation phases to design a colonoscopy non-adherence prediction model. An NLP-derived variable called the Non-Adherence Ratio ("NAR") was developed, validated, and included in the model. Patients scheduled for outpatient colonoscopy at an Academic Medical Center (AMC) that is part of a multi-hospital health system, 2009 to 2011, were included in the study. Odds ratios for non-adherence were calculated for all variables in the discovery cohort, and an Area Under the Receiver Operating Curve (AUC) was calculated for the final non-adherence prediction model. The non-adherence model included six variables: 1) gender; 2) history of psychiatric illness, 3) NAR; 4) wait time in months; 5) number of prior missed endoscopies; and 6) education level. The model achieved discrimination in the validation cohort (AUC= =70.2 %). At a threshold non-adherence score of 0.46, the model's sensitivity and specificity were 33 % and 92 %, respectively. Removing the NAR from the model significantly reduced its predictive power (AUC = 64.3 %, difference = 5.9 %, p < 0.001). A six-variable model using readily available clinical and demographic information demonstrated accuracy for predicting colonoscopy non-adherence. The NAR, a novel variable developed using NLP technology, significantly strengthened this model's predictive power.

The ability of NT-proBNP to detect chronic heart failure and predict all-cause mortality is higher in elderly Chinese coronary artery disease patients with chronic kidney disease

ObjectiveTo analyze the relationship between N-terminal pro-brain natriuretic peptide (NT-proBNP) and renal function, and compare the ability and cut-off thresholds of NT-proBNP to detect chronic heart failure (CHF) and predict mortality in elderly Chinese coronary artery disease (CAD) patients with and without chronic kidney disease (CKD).MethodsThe study included 999 CAD patients older than 60 years. The endpoint was all-cause mortality over a mean follow-up period of 417 days.ResultsThe median age was 86 years (range: 60–104 years), and the median NT-proBNP level was 409.8 pg/mL. CKD was present in 358 patients. Three hundred and six patients were positive for CHF. One hundred and ten CKD patients and 105 non-CKD patients died. Not only CKD, but also estimated glomerular filtration rate independently affected NT-proBNP. NT-proBNP detected CHF with a cut-off value of 298.4 pg/mL in non-CKD patients and a cut-off value of 435.7 pg/mL in CKD patients. NT-proBNP predicted death with a cut-off value of 369.5 pg/mL in non-CKD patients and a cut-off value of 2584.1 pg/mL in CKD patients. The NT-proBNP level was significantly related to the prevalence of CHF and all-cause mortality in CAD patients with and without CKD; this effect persisted after adjustment. The crude and multiple adjusted hazard ratios of NT-proBNP to detect CHF and predict mortality were significantly higher in patients with CKD compared with the remainder of the population. The addition of NT-proBNP to the three-variable and six-variable models generated a significant increase in the C-statistic.ConclusionAmongst elderly Chinese CAD patients, there was an independently inverse association between NT-proBNP and renal function. With the higher cutoff points, NT-proBNP better detected CHF and better predicted mortality in CKD patients than in non-CKD patients.

The Structure of Metabolic Syndrome Components Across Follow-Up Survey From Childhood to Adolescence

BackgroundThe choice of what parameters are needed for the diagnosis of Metabolic syndrome (MetS) has been criticized due to the lack of an actual “gold standard” diagnostic test even in adults. This problem seems to be greater in children and adolescents.ObjectivesStability assessment of factor structure underlying metabolic syndrome (MetS) components from childhood to adolescence in a panel study.Patients and MethodsA total number of 643 (305 boys and 338 girls) children (from 1999 to 2001), aged 6-10 years, with a complete median follow-up of 6.7 years (from 2006 to 2008) were selected among participants of Tehran Lipid and Glucose Study. We proposed 6 measured variables based on risk factors defined in Adult Treatment Panel III guidelines to describe clustering of MetS components.ResultsThe Goodness of fit of the two-factor model, extracted from exploratory factor analysis, was appropriate for boys and girls in both stages of the study using confirmatory factor analysis. Systolic blood pressure (SBP) and triglycerides (TGs), with parameter estimates (PE) of 1 and 0.75, respectively, were the greatest risk factors at baseline in boys and girls. Waist circumference with PE of 0.88 and 0.62, and SBP with PE of 0.99 and 0.86 in adolescent boys and girls, respectively, were important risk factors.ConclusionsOur panel study supports the stability of the two-factor six-variable model across two developmental stages from childhood to adolescence, among which adiposity, SBP, and TG were the predominant risk factors.

A Cardiovascular Risk Calculator for Renal Transplant Recipients

Renal transplant recipients (RTRs) have increased cardiovascular disease (CVD) risk. Standard CVD risk calculators are poorly predictive in RTRs; we therefore aimed to develop and validate an equation for CVD risk prediction in this population. We used data from the Assessment of Lescol in Renal Transplantation trial, which are randomly divided into an assessment sample and a test sample (67% and 33%, respectively, of the total population). For variable selection in the assessment sample, backward stepwise Cox regression was used. Using the regression coefficients and centralized prognostic index, risk was calculated for individual patients. The equation was then validated for calibration and discrimination using the test sample. Major adverse cardiac events could be predicted using a seven-variable model including age, previous coronary heart disease, diabetes, low-density lipoprotein, creatinine, number of transplants, and smoking. The calibration of the model was good in the test sample with a Hosmer-Lemeshow chi-square value of 11.47 and a P value of 0.245. The areas under the receiver operating characteristic curve were 0.738 in the assessment sample and 0.740 in the test sample. Total mortality could be predicted using a six-variable model including age, coronary heart disease, diabetes, creatinine, total time on renal replacement therapy, and smoking. The calibration of the model was acceptable in the test sample with a Hosmer-Lemeshow chi-square value of 13.08 and a P value of 0.109. The areas under the receiver operating characteristic curve were 0.734 in the assessment sample and 0.720 in the test sample. Using the Assessment of Lescol in Renal Transplantation trial population, a formula for 7-year CVD and mortality risk calculation for prevalent RTRs has been developed.

Multicentre validation of a prognostic index for overall survival in chronic lymphocytic leukaemia

We wish to validate in a multicentric CLL population a nomogram and a risk score recently developed to predict overall survival (OS). Complete records from 1037 CLL patients were retrospectively collected to estimate OS and time to treatment (TTT). Cox models were used to test the independence of age, β-2-microglobulin, absolute lymphocyte count (ALC), sex, Rai stage and number of involved lymph node regions (LNR). Accuracy of prognostic models was tested with the concordance index (c-index). Median follow-up was 5.5 years, with 151 deaths and 475 treated patients. Median OS was not reached (65% survival rate at 13.9 years), median TTT was 6 years. We confirmed the ability of the prognostic score to predict OS and TTT in three risk groups, with results comparable with those reported in the original report. However, ALC and Rai stage were not independent predictors, whereas the Binet staging system, which incorporates LNR variable, showed independent predictive power; furthermore, both 5- and 10-year OS estimates from nomogram were lower compared to real data. When separately analysed, the impact of therapy on OS was not selected as independent predictor of OS in our series. According to these results, we proposed a simpler four-variable model (age, sex, Binet staging, β-2-microglobulin) and a new nomogram. This model had a c-index of 0.78 versus 0.76 of the six-variable model (p = 0.043), showing better predictive accuracy. External validation and refinement are needed on independent data sets, possibly from cancer registry patients' series.

The role of granules within viscous capture threads of orb-weaving spiders

Sticky viscous prey capture threads form the spiral elements of spider orb-webs and are responsible for retaining insects that strike a web. These threads are formed of regularly spaced aqueous droplets that surround a pair of supporting axial fibers. When a thread is flattened on a microscope slide a small, opaque granule can usually be seen within each droplet. These granules have been thought to be the glycoprotein glue that imparts thread adhesion. Both independent contrast and standard regressions showed that granule size is directly related to droplet volume and indicated that granule volume is about 15% of droplet volume. We attempted to find support for the hypothesized adhesive role of granules by establishing an association between the contact surface area and volume of these granules and the stickiness of the viscous threads of 16 species in the context of a six-variable model that describes thread stickiness. However, we found that granule size made either an insignificant or a small negative contribution to thread stickiness. Consequently, we hypothesize that granules serve to anchor larger, surrounding layers of transparent glycoprotein glue to the axial fibers of the thread, thereby equipping droplets to resist slippage on the axial fibers as these droplets generate adhesion, elongate under a load, and transfer force to the axial fibers.

The adhesive delivery system of viscous capture threads spun by orb-weaving spiders

The sticky viscous capture threads in araneoid orb-webs are responsible for retaining insects that strike these webs. We used features of 16 species' threads and the stickiness that they expressed on contact plates of four widths to model their adhesive delivery systems. Our results confirm that droplets at the edges of thread contact contribute the greatest adhesion, with each successively interior droplet contributing only 0.70 as much adhesion. Thus, regardless of the size and spacing of a thread's large primary droplets, little adhesion accrues beyond a span of 20 droplets. From this pattern we computed effective droplet number (EDN), an index that describes the total droplet equivalents that contribute to the stickiness of thread spans. EDN makes the greatest positive contribution to thread stickiness, followed by an index of the shape and size of primary droplets, and the volume of small secondary droplets. The proportion of water in droplets makes the single greatest negative contribution to thread stickiness, followed by a thread's extensibility, and the area of flattened droplets. Although highly significant, this six-variable model failed to convincingly describe the stickiness of six species, a problem resolved when species were assigned to three groups and a separate model was constructed for each. These models place different weights on the variables and, in some cases, reverse or exclude the contribution of a variable. Differences in threads may adapt them to particular habitats, web architectures or prey types, or they may be shaped by a species' phylogeny or metabolic capabilities.