Multiprobabilistic prediction in early medical diagnoses

This paper describes the methodology of providing multiprobability predictions for proteomic mass spectrometry data. The methodology is based on a newly developed machine learning framework called Venn machines. Is allows to output a valid probability interval. The methodology is designed for mass spectrometry data. For demonstrative purposes, we applied this methodology to MALDI-TOF data sets in order to predict the diagnosis of heart disease and early diagnoses of ovarian cancer and breast cancer. The experiments showed that probability intervals are narrow, that is, the output of the multiprobability predictor is similar to a single probability distribution. In addition, probability intervals produced for heart disease and ovarian cancer data were more accurate than the output of corresponding probability predictor. When Venn machines were forced to make point predictions, the accuracy of such predictions is for the most data better than the accuracy of the underlying algorithm that outputs single probability distribution of a label. Application of this methodology to MALDI-TOF data sets empirically demonstrates the validity. The accuracy of the proposed method on ovarian cancer data rises from 66.7 % 11 months in advance of the moment of diagnosis to up to 90.2 % at the moment of diagnosis. The same approach has been applied to heart disease data without time dependency, although the achieved accuracy was not as high (up to 69.9 %). The methodology allowed us to confirm mass spectrometry peaks previously identified as carrying statistically significant information for discrimination between controls and cases.

A new method for the characterisation and verification of local spatial predictability for convective scale ensembles

The use of kilometre-scale ensembles in operational forecasting provides new challenges for forecast interpretation and evaluation to account for uncertainty on the convective scale. A new neighbourhood based method is presented for evaluating and characterising the local predictability variations from convective scale ensembles. Spatial scales over which ensemble forecasts agree (agreement scales, S^A) are calculated at each grid point ij, providing a map of the spatial agreement between forecasts. By comparing the average agreement scale obtained from ensemble member pairs (S^A(mm)_ij), with that between members and radar observations (S^A(mo)_ij), this approach allows the location-dependent spatial spread-skill relationship of the ensemble to be assessed. The properties of the agreement scales are demonstrated using an idealised experiment. To demonstrate the methods in an operational context the S^A(mm)_ij and S^A(mo)_ij are calculated for six convective cases run with the Met Office UK Ensemble Prediction System. The S^A(mm)_ij highlight predictability differences between cases, which can be linked to physical processes. Maps of S^A(mm)_ij are found to summarise the spatial predictability in a compact and physically meaningful manner that is useful for forecasting and for model interpretation. Comparison of S^A(mm)_ij and S^A(mo)_ij demonstrates the case-by-case and temporal variability of the spatial spread-skill, which can again be linked to physical processes.

Metacognition and proofreading: the roles of aging, motivation, and interest

Objectives: The current study examined younger and older adults’ error detection accuracy, prediction calibration, and postdiction calibration on a proofreading task, to determine if age-related difference would be present in this type of common error detection task. Method: Participants were given text passages, and were first asked to predict the percentage of errors they would detect in the passage. They then read the passage and circled errors (which varied in complexity and locality), and made postdictions regarding their performance, before repeating this with another passage and answering a comprehension test of both passages. Results: There were no age-related differences in error detection accuracy, text comprehension, or metacognitive calibration, though participants in both age groups were overconfident overall in their metacognitive judgments. Both groups gave similar ratings of motivation to complete the task. The older adults rated the passages as more interesting than younger adults did, although this level of interest did not appear to influence error-detection performance. Discussion: The age equivalence in both proofreading ability and calibration suggests that the ability to proofread text passages and the associated metacognitive monitoring used in judging one’s own performance are maintained in aging. These age-related similarities persisted when younger adults completed the proofreading tasks on a computer screen, rather than with paper and pencil. The findings provide novel insights regarding the influence that cognitive aging may have on metacognitive accuracy and text processing in an everyday task.

A device to improve the Schleger and Turner method for sweating rate measurements

The objective of this study was to test a device developed to improve the functionality, accuracy and precision of the original technique for sweating rate measurements proposed by Schleger and Turner [Schleger AV, Turner HG (1965) Aust J Agric Res 16:92-106]. A device was built for this purpose and tested against the original Schleger and Turner technique. Testing was performed by measuring sweating rates in an experiment involving six Mertolenga heifers subjected to four different thermal levels in a climatic chamber. The device exhibited no functional problems and the results obtained with its use were more consistent than with the Schleger and Turner technique. There was no difference in the reproducibility of the two techniques (same accuracy), but measurements performed with the new device had lower repeatability, corresponding to lower variability and, consequently, to higher precision. When utilizing this device, there is no need for physical contact between the operator and the animal to maintain the filter paper discs in position. This has important advantages: the animals stay quieter, and several animals can be evaluated simultaneously. This is a major advantage because it allows more measurements to be taken in a given period of time, increasing the precision of the observations and diminishing the error associated with temporal hiatus (e.g., the solar angle during field studies). The new device has higher functional versatility when taking measurements in large-scale studies (many animals) under field conditions. The results obtained in this study suggest that the technique using the device presented here could represent an advantageous alternative to the original technique described by Schleger and Turner.

Prediction of Viscosities of Fatty Compounds and Biodiesel by Group Contribution

In the present work, a group contribution method is proposed for the estimation of viscosity of fatty compounds and biodiesel esters as a function of the temperature. The databank used for regression of the group contribution parameters (1070 values for 65 types of substances) included fatty compounds, such as fatty acids, methyl and ethyl esters and alcohols, tri- and diacylglycerols, and glycerol. The inclusion of new experimental data for fatty esters, a partial acylglycerol, and glycerol allowed for a further refinement in the performance of this methodology in comparison to a prior group contribution equation (Ceriani, R.; Goncalves, C. B.; Rabelo, J.; Caruso, M.; Cunha, A. C. C.; Cavaleri, F. W.; Batista, E. A. C.; Meirelles, A. J. A. Group contribution model for predicting viscosity of fatty compounds. J. Chem. Eng. Data 2007, 52, 965-972) for all classes of fatty compounds. Besides, the influence of small concentrations of partial acylglycerols, intermediate compounds in the transesterification reaction, in the viscosity of biodiesels was also investigated.

Numerical prediction of three-dimensional time-dependent viscoelastic extrudate swell using differential and algebraic models

This study investigates the numerical simulation of three-dimensional time-dependent viscoelastic free surface flows using the Upper-Convected Maxwell (UCM) constitutive equation and an algebraic explicit model. This investigation was carried out to develop a simplified approach that can be applied to the extrudate swell problem. The relevant physics of this flow phenomenon is discussed in the paper and an algebraic model to predict the extrudate swell problem is presented. It is based on an explicit algebraic representation of the non-Newtonian extra-stress through a kinematic tensor formed with the scaled dyadic product of the velocity field. The elasticity of the fluid is governed by a single transport equation for a scalar quantity which has dimension of strain rate. Mass and momentum conservations, and the constitutive equation (UCM and algebraic model) were solved by a three-dimensional time-dependent finite difference method. The free surface of the fluid was modeled using a marker-and-cell approach. The algebraic model was validated by comparing the numerical predictions with analytic solutions for pipe flow. In comparison with the classical UCM model, one advantage of this approach is that computational workload is substantially reduced: the UCM model employs six differential equations while the algebraic model uses only one. The results showed stable flows with very large extrudate growths beyond those usually obtained with standard differential viscoelastic models. (C) 2010 Elsevier Ltd. All rights reserved.

Performance of balanced two-stage empirical predictors of realized cluster latent values from finite populations: A simulation study

Predictors of random effects are usually based on the popular mixed effects (ME) model developed under the assumption that the sample is obtained from a conceptual infinite population; such predictors are employed even when the actual population is finite. Two alternatives that incorporate the finite nature of the population are obtained from the superpopulation model proposed by Scott and Smith (1969. Estimation in multi-stage surveys. J. Amer. Statist. Assoc. 64, 830-840) or from the finite population mixed model recently proposed by Stanek and Singer (2004. Predicting random effects from finite population clustered samples with response error. J. Amer. Statist. Assoc. 99, 1119-1130). Predictors derived under the latter model with the additional assumptions that all variance components are known and that within-cluster variances are equal have smaller mean squared error (MSE) than the competitors based on either the ME or Scott and Smith`s models. As population variances are rarely known, we propose method of moment estimators to obtain empirical predictors and conduct a simulation study to evaluate their performance. The results suggest that the finite population mixed model empirical predictor is more stable than its competitors since, in terms of MSE, it is either the best or the second best and when second best, its performance lies within acceptable limits. When both cluster and unit intra-class correlation coefficients are very high (e.g., 0.95 or more), the performance of the empirical predictors derived under the three models is similar. (c) 2007 Elsevier B.V. All rights reserved.

Prediction in Multilevel Logistic Regression

The purpose of this article is to present a new method to predict the response variable of an observation in a new cluster for a multilevel logistic regression. The central idea is based on the empirical best estimator for the random effect. Two estimation methods for multilevel model are compared: penalized quasi-likelihood and Gauss-Hermite quadrature. The performance measures for the prediction of the probability for a new cluster observation of the multilevel logistic model in comparison with the usual logistic model are examined through simulations and an application.

Simultaneous prediction of symptom severity and cause in data from a test battery for Parkinson patients, using machine learning methods

The main purpose of this thesis project is to prediction of symptom severity and cause in data from test battery of the Parkinson’s disease patient, which is based on data mining. The collection of the data is from test battery on a hand in computer. We use the Chi-Square method and check which variables are important and which are not important. Then we apply different data mining techniques on our normalize data and check which technique or method gives good results.The implementation of this thesis is in WEKA. We normalize our data and then apply different methods on this data. The methods which we used are Naïve Bayes, CART and KNN. We draw the Bland Altman and Spearman’s Correlation for checking the final results and prediction of data. The Bland Altman tells how the percentage of our confident level in this data is correct and Spearman’s Correlation tells us our relationship is strong. On the basis of results and analysis we see all three methods give nearly same results. But if we see our CART (J48 Decision Tree) it gives good result of under predicted and over predicted values that’s lies between -2 to +2. The correlation between the Actual and Predicted values is 0,794in CART. Cause gives the better percentage classification result then disability because it can use two classes.

Likelihood prediction for generalized linear mixed models under covariate uncertainty

This paper presents the techniques of likelihood prediction for the generalized linear mixed models. Methods of likelihood prediction is explained through a series of examples; from a classical one to more complicated ones. The examples show, in simple cases, that the likelihood prediction (LP) coincides with already known best frequentist practice such as the best linear unbiased predictor. The paper outlines a way to deal with the covariate uncertainty while producing predictive inference. Using a Poisson error-in-variable generalized linear model, it has been shown that in complicated cases LP produces better results than already know methods.

Prediction Of Hydrological Models’ Uncertainty By A Committee Of Machine Learning-Models

This study presents an approach to combine uncertainties of the hydrological model outputs predicted from a number of machine learning models. The machine learning based uncertainty prediction approach is very useful for estimation of hydrological models' uncertainty in particular hydro-metrological situation in real-time application [1]. In this approach the hydrological model realizations from Monte Carlo simulations are used to build different machine learning uncertainty models to predict uncertainty (quantiles of pdf) of the a deterministic output from hydrological model . Uncertainty models are trained using antecedent precipitation and streamflows as inputs. The trained models are then employed to predict the model output uncertainty which is specific for the new input data. We used three machine learning models namely artificial neural networks, model tree, locally weighted regression to predict output uncertainties. These three models produce similar verification results, which can be improved by merging their outputs dynamically. We propose an approach to form a committee of the three models to combine their outputs. The approach is applied to estimate uncertainty of streamflows simulation from a conceptual hydrological model in the Brue catchment in UK and the Bagmati catchment in Nepal. The verification results show that merged output is better than an individual model output. [1] D. L. Shrestha, N. Kayastha, and D. P. Solomatine, and R. Price. Encapsulation of parameteric uncertainty statistics by various predictive machine learning models: MLUE method, Journal of Hydroinformatic, in press, 2013.

A Gradient-Type Method For Real-Time State Estimation Of Water Distribution Networks

Drinking water distribution networks risk exposure to malicious or accidental contamination. Several levels of responses are conceivable. One of them consists to install a sensor network to monitor the system on real time. Once a contamination has been detected, this is also important to take appropriate counter-measures. In the SMaRT-OnlineWDN project, this relies on modeling to predict both hydraulics and water quality. An online model use makes identification of the contaminant source and simulation of the contaminated area possible. The objective of this paper is to present SMaRT-OnlineWDN experience and research results for hydraulic state estimation with sampling frequency of few minutes. A least squares problem with bound constraints is formulated to adjust demand class coefficient to best fit the observed values at a given time. The criterion is a Huber function to limit the influence of outliers. A Tikhonov regularization is introduced for consideration of prior information on the parameter vector. Then the Levenberg-Marquardt algorithm is applied that use derivative information for limiting the number of iterations. Confidence intervals for the state prediction are also given. The results are presented and discussed on real networks in France and Germany.

Novo método para estimativa do gradiente de fratura para poços de petróleo

The development of oil wells drilling requires additional cares mainly if the drilling is in offshore ultra deep water with low overburden pressure gradients which cause low fracture gradients and, consequently, difficult the well drilling by the reduction of the operational window. To minimize, in the well planning phases, the difficulties faced by the drilling in those sceneries, indirect models are used to estimate fracture gradient that foresees approximate values for leakoff tests. These models generate curves of geopressures that allow detailed analysis of the pressure behavior for the whole well. Most of these models are based on the Terzaghi equation, just differentiating in the determination of the values of rock tension coefficient. This work proposes an alternative method for prediction of fracture pressure gradient based on a geometric correlation that relates the pressure gradients proportionally for a given depth and extrapolates it for the whole well depth, meaning that theses parameters vary in a fixed proportion. The model is based on the application of analytical proportion segments corresponding to the differential pressure related to the rock tension. The study shows that the proposed analytical proportion segments reaches values of fracture gradient with good agreement with those available for leakoff tests in the field area. The obtained results were compared with twelve different indirect models for fracture pressure gradient prediction based on the compacting effect. For this, a software was developed using Matlab language. The comparison was also made varying the water depth from zero (onshore wellbores) to 1500 meters. The leakoff tests are also used to compare the different methods including the one proposed in this work. The presented work gives good results for error analysis compared to other methods and, due to its simplicity, justify its possible application

Modelos matemáticos para estimativa do consumo máximo de oxigênio pela ventilometria de esforço em indivíduos saudáveis

The relation between metabolic demand and maximal oxygen consumption during exercise have been investigated in different areas of knowledge. In the health field, the determination of maximal oxygen consumption (VO2max) is considered a method to classify the level of physical fitness or the risk of cardiocirculatory diseases. The accuracy to obtain data provides a better evaluation of functional responses and allows a reduction in the error margin at the moment of risk classification, as well as, at the moment of determination of aerobic exercise work load. In Brasil, the use of respirometry associated to ergometric test became an opition in the cardiorespiratory evaluation. This equipment allows predictions concerning the oxyredutase process, making it possible to identify physiological responses to physical effort as the respiratory threshold. This thesis focused in the development of mathematical models developed by multiple regression validated by the stepwise method, aiming to predict the VO2max based on respiratory responses to physical effort. The sample was composed of a ramdom sample of 181 healthy individuals, men and women, that were randomized to two groups: regression group and cross validation group (GV). The voluntiars were submitted to a incremental treadmill test; objetiving to determinate of the second respiratory threshold (LVII) and the Peak VO2max. Using the método forward addition method 11 models of VO2max prediction in trendmill were developded. No significative differences were found between the VO2max meansured and the predicted by models when they were compared using ANOVA One-Way and the Post Hoc test of Turkey. We concluded that the developed mathematical models allow a prediction of the VO2max of healthy young individuals based on the LVII

Leaf area prediction models for Zinnia elegans Jacq., Zinnia haageana regel and 'profusion cherry'

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)