Comparison of machine learning approaches and statistical methods for estimation of airborne chemical exposures

Accurate quantitative estimation of exposure using retrospective data has been one of the most challenging tasks in the exposure assessment field. To improve these estimates, some models have been developed using published exposure databases with their corresponding exposure determinants. These models are designed to be applied to reported exposure determinants obtained from study subjects or exposure levels assigned by an industrial hygienist, so quantitative exposure estimates can be obtained. ^ In an effort to improve the prediction accuracy and generalizability of these models, and taking into account that the limitations encountered in previous studies might be due to limitations in the applicability of traditional statistical methods and concepts, the use of computer science- derived data analysis methods, predominantly machine learning approaches, were proposed and explored in this study. ^ The goal of this study was to develop a set of models using decision trees/ensemble and neural networks methods to predict occupational outcomes based on literature-derived databases, and compare, using cross-validation and data splitting techniques, the resulting prediction capacity to that of traditional regression models. Two cases were addressed: the categorical case, where the exposure level was measured as an exposure rating following the American Industrial Hygiene Association guidelines and the continuous case, where the result of the exposure is expressed as a concentration value. Previously developed literature-based exposure databases for 1,1,1 trichloroethane, methylene dichloride and, trichloroethylene were used. ^ When compared to regression estimations, results showed better accuracy of decision trees/ensemble techniques for the categorical case while neural networks were better for estimation of continuous exposure values. Overrepresentation of classes and overfitting were the main causes for poor neural network performance and accuracy. Estimations based on literature-based databases using machine learning techniques might provide an advantage when they are applied to other methodologies that combine `expert inputs' with current exposure measurements, like the Bayesian Decision Analysis tool. The use of machine learning techniques to more accurately estimate exposures from literature-based exposure databases might represent the starting point for the independence from the expert judgment.^

Automatic blood glucose classification for gestational diabetes with feature selection: decision trees vs neural networks

Automatic blood glucose classification may help specialists to provide a better interpretation of blood glucose data, downloaded directly from patients glucose meter and will contribute in the development of decision support systems for gestational diabetes. This paper presents an automatic blood glucose classifier for gestational diabetes that compares 6 different feature selection methods for two machine learning algorithms: neural networks and decision trees. Three searching algorithms, Greedy, Best First and Genetic, were combined with two different evaluators, CSF and Wrapper, for the feature selection. The study has been made with 6080 blood glucose measurements from 25 patients. Decision trees with a feature set selected with the Wrapper evaluator and the Best first search algorithm obtained the best accuracy: 95.92%.

Monitoring dendritic cells in clinical practice using a new whole blood single-platform TruCOUNT assay

Dendritic cells (DC) from distinct DC subsets are essential contributors to normal human immune responses. Despite this, reliable assays that enable DC to be counted precisely have been slow to evolve. We have now developed a new single-platform flow cytometric assay based on TruCOUN(TM) beads and the whole blood Lyse/No-Wash protocol that allows precise counting of the CD14(-) blood DC subsets: CD11c(+)CD16(-) DC, CD11c(+)CD16(+) DC, CD123(hi) DC, CD1c(+) DC and BDCA-3(+) DC. This assay requires 50 mul of whole blood; does not rely on a hematology blood analyser for the absolute DC counts; allows DC counting in EDTA samples 24 It after collection; and is suitable for cord blood and peripheral blood. The data is highly reproducible with intra-assay and inter-assay coefficients of variation less than 3% and 11%, respectively. This assay does not produce the DC-T lymphocyte conjugates that result in DC counting abnormalities in conventional gradient-density separation procedures. Using the TruCOUNT assay, we established that absolute blood DC counts reduce with age in healthy individuals. In preliminary studies, we found a significantly lower absolute blood CD11c(+)CD16(+) DC count in stage III/IV versus stage I/II breast carcinoma patients and a lower absolute blood CD123(hi) DC count in multiple myeloma patients, compared to age-matched controls. These data indicate that scientific progress in DC counting technology will lead to the global standardization of DC counting and allow clinically meaningful data to be obtained. (C) 2003 Elsevier B.V. All rights reserved.

Serum antibodies to oral anaerobic bacteria in patients with rheumatoid arthritis

BACKGROUND: This study was conducted to determine the component that causes the disease in rheumatoid arthritis (RA), which shows great resemblance to periodontitis in a pathologic context. MATERIALS AND METHODS: Within this study, the pathogen-specific IgG levels formed against Porphyromonas gingivalis FDC 381, Prevotella melaninogenica ATCC 25845, Actinobacillus actinomycetemcomitans Y4, Bacteroides forsythus ATCC 43047, and Prevotella intermedia 25611 oral bacteria were researched from the blood serum samples of 30 RA patients and 20 healthy controls with the enzyme-linked immunosorbent assay (ELISA) method. RESULTS: The IgG levels of P gingivalis, P intermedia, P melaninogenica, and B forsythus were found to be significantly higher in RA patients when compared with those of the controls. Of the other bacteria antibodies, A actinomycetemcomitans was not found at greater levels in RA serum samples in comparison with the healthy samples. CONCLUSION: The antibodies formed against P gingivalis, P intermedia, P melaninogenica, and B forsythus could be important to the etiopathogenesis of RA.

Novel visualization methods for protein data

Visualization of high-dimensional data has always been a challenging task. Here we discuss and propose variants of non-linear data projection methods (Generative Topographic Mapping (GTM) and GTM with simultaneous feature saliency (GTM-FS)) that are adapted to be effective on very high-dimensional data. The adaptations use log space values at certain steps of the Expectation Maximization (EM) algorithm and during the visualization process. We have tested the proposed algorithms by visualizing electrostatic potential data for Major Histocompatibility Complex (MHC) class-I proteins. The experiments show that the variation in the original version of GTM and GTM-FS worked successfully with data of more than 2000 dimensions and we compare the results with other linear/nonlinear projection methods: Principal Component Analysis (PCA), Neuroscale (NSC) and Gaussian Process Latent Variable Model (GPLVM).

Using secondary knowledge to support decision tree classification of retrospective clinical data

Retrospective clinical data presents many challenges for data mining and machine learning. The transcription of patient records from paper charts and subsequent manipulation of data often results in high volumes of noise as well as a loss of other important information. In addition, such datasets often fail to represent expert medical knowledge and reasoning in any explicit manner. In this research we describe applying data mining methods to retrospective clinical data to build a prediction model for asthma exacerbation severity for pediatric patients in the emergency department. Difficulties in building such a model forced us to investigate alternative strategies for analyzing and processing retrospective data. This paper describes this process together with an approach to mining retrospective clinical data by incorporating formalized external expert knowledge (secondary knowledge sources) into the classification task. This knowledge is used to partition the data into a number of coherent sets, where each set is explicitly described in terms of the secondary knowledge source. Instances from each set are then classified in a manner appropriate for the characteristics of the particular set. We present our methodology and outline a set of experiential results that demonstrate some advantages and some limitations of our approach. © 2008 Springer-Verlag Berlin Heidelberg.

Statistical methods for polyhedral shape classification with incomplete data - application to cryo-electron tomographic images

A certain type of bacterial inclusion, known as a bacterial microcompartment, was recently identified and imaged through cryo-electron tomography. A reconstructed 3D object from single-axis limited angle tilt-series cryo-electron tomography contains missing regions and this problem is known as the missing wedge problem. Due to missing regions on the reconstructed images, analyzing their 3D structures is a challenging problem. The existing methods overcome this problem by aligning and averaging several similar shaped objects. These schemes work well if the objects are symmetric and several objects with almost similar shapes and sizes are available. Since the bacterial inclusions studied here are not symmetric, are deformed, and show a wide range of shapes and sizes, the existing approaches are not appropriate. This research develops new statistical methods for analyzing geometric properties, such as volume, symmetry, aspect ratio, polyhedral structures etc., of these bacterial inclusions in presence of missing data. These methods work with deformed and non-symmetric varied shaped objects and do not necessitate multiple objects for handling the missing wedge problem. The developed methods and contributions include: (a) an improved method for manual image segmentation, (b) a new approach to 'complete' the segmented and reconstructed incomplete 3D images, (c) a polyhedral structural distance model to predict the polyhedral shapes of these microstructures, (d) a new shape descriptor for polyhedral shapes, named as polyhedron profile statistic, and (e) the Bayes classifier, linear discriminant analysis and support vector machine based classifiers for supervised incomplete polyhedral shape classification. Finally, the predicted 3D shapes for these bacterial microstructures belong to the Johnson solids family, and these shapes along with their other geometric properties are important for better understanding of their chemical and biological characteristics.

Adiposity, Sex Hormones, and Repetitive Element DNA Methylation in Healthy Postmenopausal Women

INTRODUCTION: Low levels of methylation within repetitive DNA elements, such as long interspersed nuclear element-1 (LINE-1) and Alu repeats, are believed to epigenetically predispose an individual to cancer and other diseases. The extent to which lifestyle factors affect the degree of DNA methylation within these genomic regions has yet to be fully understood. Adiposity and sex hormones are established risk factors for certain types of cancer and other illnesses, particularly amongst postmenopausal women. The aim of the current investigation is to assess the impact of adiposity and sex hormones on LINE-1 and Alu methylation in healthy postmenopausal women. METHODS: A cross-sectional study was conducted using baseline data from an ancillary study of the Alberta Physical Activity and Breast Cancer Prevention (ALPHA) Trial. Current adiposity was measured using a dual-energy x-ray absorptiometry (DXA) scan, computed tomography (CT) scan, and balance beam scale. Historical weights were self-reported in a questionnaire. Current endogenous sex hormone concentrations were measured in fasting blood serum. Estimated lifetime number of menstrual cycles was used as a proxy for cumulative exposure to ovarian sex hormones. Repetitive element methylation was quantified in white blood cells using a pyrosequencing assay. Linear regression was used to model the relations of interest while adjusting for important confounders. RESULTS: Adiposity and serum estrogen concentrations were positively related to LINE-1 methylation but were not associated with Alu methylation. Cumulative ovarian sex hormone exposure had a “U-shaped” relation with LINE-1 regardless of folate intake and a negative relation with Alu methylation amongst low folate consumers. Androgens were not associated with repetitive element DNA methylation in this population. CONCLUSION: Adiposity and estrogens appear to play a role in maintaining high levels of repetitive element DNA methylation in healthy postmenopausal women. LINE-1 methylation may be a mechanism whereby estrogen exposure protects against cardiovascular and neurodegenerative illnesses. These results add to the growing body of literature showing how the epigenome is shaped by our lifestyle choices. Future prospective studies assessing the relation between levels of repetitive element DNA methylation in healthy individuals and subsequent disease risk are needed to better understand the clinical significance of these results.

A comparison of different pre-lysis methods and extraction kits for recovery of Stretococcus agalacticae (Lancefield group B Streptococcus) DNA from whole blood

Sub-optimal recovery of bacterial DNA from whole blood samples can limit the sensitivity of molecular assays to detect pathogenic bacteria. We compared 3 different pre-lysis protocols (none, mechanical pre-lysis and achromopeptidasepre-lysis) and 5 commercially available DNA extraction platforms for direct detection of Group B Streptococcus (GBS) in spiked whole blood samples, without enrichment culture. DNA was extracted using the QIAamp Blood Mini kit (Qiagen), UCP Pathogen Mini kit (Qiagen), QuickGene DNA Whole Blood kit S (Fuji), Speed Xtract Nucleic Acid Kit 200 (Qiagen) and MagNA Pure Compact Nucleic Acid Isolation Kit I (Roche Diagnostics Corp). Mechanical pre-lysis increased yields of bacterial genomic DNA by 51.3 fold (95% confidence interval; 31.6–85.1, p < 0.001) and pre-lysis with achromopeptidase by 6.1 fold (95% CI; 4.2–8.9, p < 0.001), compared with no pre-lysis. Differences in yield dueto pre-lysis were 2–3 fold larger than differences in yield between extraction methods. Including a pre-lysis step can improve the limits of detection of GBS using PCR or other molecular methods without need for culture.

The impact of serum uric acid on the diagnostic of metabolic syndrome in apparently healthy Brazilian middle-aged men

Background: Hyperuricemia is related to Metabolic Syndrome (MetS) and cardiovascular diseases, but the use of serum uric acid (UA) to diagnose MetS is currently ignored in clinical practices. Objectives: To examine the impact of serum UA on the diagnostic of MetS and the relationship of serum UA with cardiometabolic risk factors in apparently healthy Brazilian middle-aged men residents in a city of Minas Gerais. Methods: In a cross-sectional analysis, 289 apparently healthy middle-aged men underwent anthropometric, clinical, sociodemographic and blood serum biochemical evaluation. By using receive operating curve the internal cutoff of serum UA was determined (5.25 mg/dL). Results: Subjects with two or more components of MetS exhibited higher serum UA as compared to those with one or none component. The inclusion of serum UA ≥ 5.25mg/dL as an additional component of MetS increased the occurrence of this syndrome by 13%. Subjects with UA ≥ 5.25mg/dL showed high prevalence for MetS and association with its components (central obesity, hypertriglyceridemia, dyslipidemia and hypertension) as well as atherogenic risk. Conclusions: Serum UA has an important impact on the diagnostic of MetS and is related to cardiometabolic risk factors in apparently healthy Brazilian middle-aged men. Its use in clinical practices could aggregate accuracy to diagnose MetS.

An investigation of imputation methods for discrete databases and multi-variate time series

When it comes to information sets in real life, often pieces of the whole set may not be available. This problem can find its origin in various reasons, describing therefore different patterns. In the literature, this problem is known as Missing Data. This issue can be fixed in various ways, from not taking into consideration incomplete observations, to guessing what those values originally were, or just ignoring the fact that some values are missing. The methods used to estimate missing data are called Imputation Methods. The work presented in this thesis has two main goals. The first one is to determine whether any kind of interactions exists between Missing Data, Imputation Methods and Supervised Classification algorithms, when they are applied together. For this first problem we consider a scenario in which the databases used are discrete, understanding discrete as that it is assumed that there is no relation between observations. These datasets underwent processes involving different combina- tions of the three components mentioned. The outcome showed that the missing data pattern strongly influences the outcome produced by a classifier. Also, in some of the cases, the complex imputation techniques investigated in the thesis were able to obtain better results than simple ones. The second goal of this work is to propose a new imputation strategy, but this time we constrain the specifications of the previous problem to a special kind of datasets, the multivariate Time Series. We designed new imputation techniques for this particular domain, and combined them with some of the contrasted strategies tested in the pre- vious chapter of this thesis. The time series also were subjected to processes involving missing data and imputation to finally propose an overall better imputation method. In the final chapter of this work, a real-world example is presented, describing a wa- ter quality prediction problem. The databases that characterized this problem had their own original latent values, which provides a real-world benchmark to test the algorithms developed in this thesis.

Aplicação de métodos de calibração multivariada para a determinação simultânea de riboflavina (VB2), tiamina (VB1), piridoxina (VB6) e nicotinamida (VPP)

In this work, the artificial neural networks (ANN) and partial least squares (PLS) regression were applied to UV spectral data for quantitative determination of thiamin hydrochloride (VB1), riboflavin phosphate (VB2), pyridoxine hydrochloride (VB6) and nicotinamide (VPP) in pharmaceutical samples. For calibration purposes, commercial samples in 0.2 mol L-1 acetate buffer (pH 4.0) were employed as standards. The concentration ranges used in the calibration step were: 0.1 - 7.5 mg L-1 for VB1, 0.1 - 3.0 mg L-1 for VB2, 0.1 - 3.0 mg L-1 for VB6 and 0.4 - 30.0 mg L-1 for VPP. From the results it is possible to verify that both methods can be successfully applied for these determinations. The similar error values were obtained by using neural network or PLS methods. The proposed methodology is simple, rapid and can be easily used in quality control laboratories.