Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistical Regression”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,639 records · Page 91Linked to original sources
PubMed1979

Evaluation of response criteria in advanced lung cancer.

Because we found it illogical to attempt measurement of nonmeasurable but visible tumors in patients with advanced lung cancer, we devised and used a separate set of response criteria for patients with evaluable, nonmeasurable tumors. Patients with evaluable disease who had obtained a tumor regression after therapy were compared to patients with measurable disease who had obtained a tumor regression according to standard criteria. Among 191 cases evaluated (54% with measurable and 46% with evaluable disease), 59 tumor regressions were found (35 in patients with measurable disease and 24 in patients with evaluable disease). Evaluating regressions in patients with both types of disease, we could not detect statistical differences in regression rates, times to regression, durations of regression or survival. We believe the response criteria used for patients with evaluable, nonmeasurable lung cancer to be valid in assessing response to therapy.

Antineoplastic Agents↗
PubMed2006

Understanding community structure: a data-driven multivariate approach.

Habitat is known to influence community structure yet, because these effects are complex, elucidating these relationships has proven difficult. Multiple aspects of vegetation architecture or plant species composition, for example, may simultaneously affect animal communities and their constituent species. Many traditional statistical approaches (e.g., regression) have difficulty in handling large numbers of collinear variables. On the other hand, multivariate methods, such as ordination, are well suited to handle these large datasets, but they have primarily been used in ecology as descriptive techniques, and less frequently as a data reduction tool for predictor variables in regression. Here, I employ a multivariate approach for variable reduction of both the predictor and response variables to investigate the influences of vegetation architecture and plant species on community composition in spiders using multiple regression. This allows retention of the information in the original dataset while producing statistically tractable variables for use in further analyses. I used nonmetric multidimensional scaling to reduce the number of variables for predictor (habitat architecture and plant species) and response (spider species) data matrices, and used these new variables in multiple regression analyses. These axes can be interpreted based on their correlations with the original variables, allowing for recovery of biologically meaningful information from regressions. Consequently, the important variables are determined by the data themselves, rather than by a priori assumptions of the researcher. Contrary to expectations based on previous work in spiders and other animals, plant species composition explained more variation in spider communities than did habitat architecture, and was also a stronger predictor of other community structure variables (overall abundance, species richness, and species diversity). I discuss possible ecological explanations for these results, and the advantages of the proposed method.

Animals↗
PubMed1992

Factors involved in burn mortality: a multivariate statistical approach based on discriminant analysis.

This article suggests an alternative statistical model for studying the mortality of burned patients: discriminant analysis. This model was applied to our population of 532 patients among whom 71 died. It is not the first time that this model has been applied to assess burn mortality, although it is not frequently used for that application. We found four factors that are statistically significant: age, TBSA, inhalation injury and sex (female). Discriminant analysis allowed us to demonstrate an impressive correlation between death, age and TBSA; inhalation injury by itself and sex, the two other significant factors in our study, seem to have a minor influence on the final outcome of the burned patients and their predictive value is virtually nil. The advantages of this statistical model are compared with logistic regression, the commonly chosen statistical method.

Adult↗
PubMed2005

Influence of surface and heat treatments on the flexural strength of Y-TZP dental ceramic.

OBJECTIVE: The aim of the investigation was to assess the influence of sandblasting, grinding, grinding orientation, polishing and heat treatment on the flexural strength of a yittria stabilized tetragonal zirconia polycrystals ceramic (Y-TPZ). METHODS: The specimens (160 beams) were equally divided into four groups according to the surface treatment (sandblasted, polished, ground parallel to the tensile axis, ground perpendicular). Twenty specimens from each group underwent heat treatment under the firing conditions used to fire a layer of porcelain and glaze. After treatment, the three-point flexure test was used to calculate the flexural strength and X-ray diffraction analysis was used to estimate the relative amount of monoclinic phase. The reliability of strength was assessed through the Weibull distribution. Statistical analysis was conducted with multiple regression analysis, one-way ANOVA and Tukey's pairwise multiple comparisons. Treated and fractured surfaces were observed with SEM. RESULTS: The following values of strength and relative content of monoclinic phase of zirconia were measured for each group: sandblasted (1540MPa; 9.5%); ground parallel (1330MPa; 8.3%); ground perpendicular (1525MPa; 8.3%); ground parallel and heated (1225MPa; monoclinic content not detectable); ground perpendicular and heated (1185MPa; monoclinic content not detectable); polished and heated (1165MPa; monoclinic content not detectable); polished (1095MPa; 0.8%); sandblasted and heated (955MPa; 0.3%). CONCLUSIONS: The present study suggests that sandblasting and grinding may be recommended to increase the strength of dental Y-TZP, provided they are not followed by heat treatment. Fine polishing may remove the layer of compressive stresses and therefore, lower the mean flexural strength.

Analysis of Variance↗
PubMed2003

Assessment of local changes of cerebral perfusion and blood concentration by ultrasound harmonic B-mode contrast measurement in piglet.

This study tested the hypothesis that changes in the blood concentration, and possibly in the perfusion, of different areas in the brain can be assessed by the use of ultrasound contrast agent (CA) and (linear) echo densitometry. The experiments were performed with piglets (n=3) under general anesthesia and artificial ventilation. Ultrasound CA was administered through a femoral vein as a short bolus. First passage wash-in curve was measured from image gray level during continuous low level (mechanical index<0.2) ultrasound imaging. This curve was obtained from 1-cm2 areas of the cortex (surface), the brain stem (inner) and the left carotid artery (vessel). Cerebral hemoglobin concentration changes were measured with near-infrared spectroscopy (NIRS). This approach enabled a cross-validation of these techniques. The measurements were repeated under conditions of normocapnia, mild hypercapnia and deep hypercapnia. Several physiologic signals, as well as the carotid blood flow, were measured simultaneously and related to gray level by linear regression analysis. The most significant results found were a high R2-statistic of the regression of the percentage change of the peak of the surface and inner wash-in curves with the arterial carbon dioxide pressure (R2=0.63 and R2=0.70, respectively), the blood pH (R2=0.79 and R2=0.81), the carotid flow (R2=0.75 and R2=0.72) and the partial arterial oxygen pressure (R2=0.47 and R2=0.55). Finally, a high correlation of peak gray level with total hemoglobin concentration change, independently measured by NIRS, was found (R2=0.69). In conclusion, these experiments show a reasonable intersubject variability of various relative measures derived from gray level ultrasound wash-in curves. High sensitivity to physiologic changes related to hypercapnia was observed for the peak contrast of wash-in curves. For up-slope and area-under-the-curve (first passage) this was lower but still highly significant. The gray-level ultrasound measures are highly correlated to changes in regional hemoglobin concentration in brain tissue assessed by NIRS.

Animals↗
PubMedDate not supplied

Individual aging and mortality rate: how are they related?

Many researchers working in the area of aging and longevity base their conclusions on the behavior of empirical age trajectories of mortality rates. In such analyses, changes in the slope of the logarithm of the mortality curve are often associated with changes in the rate of individual aging. We show that such interpretation may be incorrect: the changes in the slope of this curve do not necessarily correspond to the changes in the rate of individual aging. We use three models of mortality and aging to illustrate this statement. The first one is based on the idea of frailty. We show that changes in frailty distribution alone may be responsible for changes in the slope. The second model exploits the idea of saving lives. It evaluates changes in mortality rate after elimination of lethal stressful events. The third model uses the idea of Strehler and Mildvan (1960). It shows that changes in the rate of individual aging may take place without changes in the slope of the logarithm of the mortality curve.

Aging↗
PubMed1997

Reactivity with a specific epitope of outer surface protein A predicts protection from infection with the Lyme disease spirochete, Borrelia burgdorferi.

The response to recombinant vaccines for Lyme disease was studied to determine serum antibody levels effective in protecting against tick-transmitted infection. Data presented here demonstrate a significant correlation between antibody to an epitope on outer surface protein A (OspA) and protection against infection with Borrelia burgdorferi in canines and mice. A competitive enzyme-linked immunosorbent assay was developed to measure antibody to a site on OspA, defined by monoclonal antibody LA-2. Comparison of LA-2 titers against infection of canines and mice following vaccination and challenge established a predicted value for LA-2 titers. The statistical relationship between serum antibody levels and protection was calculated by logistic regression analysis. The statistical model predicted that an LA-2 titer of 0.32 microg equivalents (eq) per ml correlated to an 80% predicted probability of protection for both mice and dogs. This value was used to classify mice and dogs as to their protected status at the time of tick exposure. The LA-2 cutoff titer (0.32 microg eq/ml) correctly classified all dogs (n = 13) and mice (n = 44) that failed to become infected. By contrast, 20 of 22 dogs and 28 of 31 mice with titers of less than 0.32 microg eq/ml became infected. On the basis of these results, we conclude that an LA-2 titer is a reliable indicator of immune status for estimating immune protection following use of OspA-based vaccines for B. burgdorferi sensu stricto.

Animals↗
PubMed2004

Gaussian processes for machine learning.

Gaussian processes (GPs) are natural generalisations of multivariate Gaussian random variables to infinite (countably or continuous) index sets. GPs have been applied in a large number of fields to a diverse range of ends, and very many deep theoretical analyses of various properties are available. This paper gives an introduction to Gaussian processes on a fairly elementary level with special emphasis on characteristics relevant in machine learning. It draws explicit connections to branches such as spline smoothing models and support vector machines in which similar ideas have been investigated. Gaussian process models are routinely used to solve hard machine learning problems. They are attractive because of their flexible non-parametric nature and computational simplicity. Treated within a Bayesian framework, very powerful statistical methods can be implemented which offer valid estimates of uncertainties in our predictions and generic model selection procedures cast as nonlinear optimization problems. Their main drawback of heavy computational scaling has recently been alleviated by the introduction of generic sparse approximations.13,78,31 The mathematical literature on GPs is large and often uses deep concepts which are not required to fully understand most machine learning applications. In this tutorial paper, we aim to present characteristics of GPs relevant to machine learning and to show up precise connections to other "kernel machines" popular in the community. Our focus is on a simple presentation, but references to more detailed sources are provided.

Algorithms↗
PubMed2005

Partial cross-correlation analysis resolves ambiguity in the encoding of multiple movement features.

A classical question in neuroscience is which features of a stimulus or of an action are represented in brain activity. When several features are interdependent either at a given point in time or at distinct points in time, neural activity related to one feature appears to be correlated with other features. Thus techniques that simultaneously consider multiple features cannot account for delayed interdependencies between features. The result is an ambiguity with respect to the encoded features. Here, we resolve this ambiguity by applying a novel statistical method based on partial cross-correlations. The method yields estimates of linear correlations between neural activity and a given feature that are not affected by linear correlations with other features at multiple time delays. The method also provides a graphical output measured on a scale that allows for comparisons between different features, neurons, and experiments. We use real movement data and neural activity simulated according to a wide range of tuning models to illustrate the method. When applied to real neural activity, the procedure yields results that indicate which of the considered features the neural activity is related to and at what time delays.

Action Potentials↗
PubMed2003

Predictors of substance abuse treatment outcomes in Tennessee.

In planning and implementing programs to treat substance abuse, it is important to understand which factors influence post-treatment abstinence. This article identifies and analyzes several variables important in predicting the likelihood of abstinence among substance abuse clients. The data used in this study was collected from 1,350 clients treated for alcohol or drug abuse in residential, halfway house, or outpatient facilities in Tennessee. We analyzed 22 variables as possible treatment outcome predictors by using two statistical procedures: stepwise logistic regression analysis and Quick, Unbiased, Efficient, Statistical Tree (QUEST) analysis, a tree-structured classification algorithm analysis. We found one pretreatment, five in-treatment, and three post-treatment variables to be significant predictors of treatment outcome: previous treatment history, perceived helpfulness of the treatment, simultaneous treatment for mental health, number of days in treatment,completion of treatment, special skills training during treatment, obtaining healthcare services for major physical health problem after treatment, living with someone using alcohol or drugs post treatment, and arrest record since treatment.

Adolescent↗
PubMed2002

Statistical methods.

The purpose of this chapter is to present a conceptual framework that applies to almost all statistical procedures discussed so far in this text. We also describe briefly, using a conceptual framework only, some of the more advanced techniques used in medicine.

Humans↗
PubMed2004

Water quality modelling of Jadro spring.

Management of water quality in karst is a specific problem. Water generally moves very fast by infiltration processes but far more by concentrated flows through fissures and openings in karst. This enables the entire surface pollution to be transferred fast and without filtration into groundwater springs. A typical example is the Jadro spring. Changes in water quality at the spring are sudden, but short. Turbidity as a major water quality problem for the karst springs regularly exceeds allowable standards. Former practice in problem solving has been reduced to intensive water disinfection in periods of great turbidity without analyses of disinfection by-products risks for water users. The main prerequisite for water quality control and an optimization of water disinfection is the knowledge of raw water quality and nature of occurrence. The analysis of monitoring data and their functional relationship with hydrological parameters enables establishment of a stochastic model that will help obtain better information on turbidity in different periods of the year. Using the model a great number of average monthly and extreme daily values are generated. By statistical analyses of these data possibility of occurrence of high turbidity in certain months is obtained. This information can be used for designing expert system for water quality management of karst springs. Thus, the time series model becomes a valuable tool in management of drinking water quality of the Jadro spring.

Disinfection↗