Rafael Izbicki

Depressive morbidity and gender in community-dwelling Brazilian elderly: systematic review and meta-analysis.

BACKGROUND Although studies indicate that community-dwelling elderly have a lower prevalence of major depression compared with younger age groups, prevalence estimates in Brazil show that clinically significant depressive symptoms (CSDS) and depression are frequent in the older population. However, a systematic review and meta-analysis of prevalence of and factors associated with depressive disorders and symptoms in elderly Brazilians has not previously been reported. The aims were (i) to perform a survey of studies dating from 1991 to 2009 on the prevalence of depressive disorders and CSDS in elderly Brazilians residing in the community; (ii) to determine depression prevalence and identify associated factors; and (iii) develop a meta-analysis to indicate the combined prevalence and the influence of gender on depressive morbidity in this population. METHODS Studies were selected from articles dated between January 1991 and May 2009, extracted from Medline, LILACS and SciELO databases. RESULTS A total of 17 studies were found, 13 with CSDS, 1 with major depression alone and 3 with major depression and dysthymia, involving the evaluation of 15,491 elderly people. The average age of participants varied between 66.5 and 84.0 years. Prevalence rates of 7.0% for major depression, 26.0% for CSDS, and 3.3% for dysthymia were found. The odds ratios for major depression and CSDS were greater among women. There was a significant association between major depression or CSDS and cardiovascular diseases. CONCLUSION The review indicates greater prevalence of both major depression and CSDS compared to rates reported in the international literature, while the prevalence of dysthymia was found to be similar. The high prevalence of CSDS and its significant association with cardiovascular diseases reinforces the importance of evaluating subthreshold depressive symptoms in the elderly in the community.

New image statistics for detecting disturbed galaxy morphologies at high redshift

Testing theories of hierarchical structure formation requires estimating the distribution of galaxy morphologies and its change with redshift. One aspect of this investigation involves identifying galaxies with disturbed morphologies (e.g. merging galaxies). This is often done by summarizing galaxy images using, e.g. the concentration, asymmetry and clumpiness and Gini-M20 statistics of Conselice and Lotz et al., respectively, and associating particular statistic values with disturbance. We introduce three statistics that enhance detection of disturbed morphologies at high redshift (z ∼ 2): the multimode (M), intensity (I) and deviation (D) statistics. We show their effectiveness by training a machine-learning classifier, random forest, using 1639 galaxies observed in the H band by the Hubble Space Telescope WFC3, galaxies that had been previously classified by eye by the Cosmic Assembly Near-IR Deep Extragalactic Legacy Survey collaboration. We find that the MID statistics (and the A statistic of Conselice) are the most useful for identifying disturbed morphologies. We also explore whether human annotators are useful for identifying disturbed morphologies. We demonstrate that they show limited ability to detect disturbance at high redshift, and that increasing their number beyond ≈10 does not provably yield better classification performance. We propose a simulation-based model-fitting algorithm that mitigates these issues by bypassing annotation.

Optimizing the CAMCOG test in the screening for mild cognitive impairment and incipient dementia: saving time with relevant domains.

To identify the CAMCOG sub‐items that best contribute for the identification of patients with mild cognitive impairment (MCI) and incipient Alzheimers disease (AD) in clinical practice.

Can the CAMCOG be a good cognitive test for patients with Alzheimer's disease with low levels of education?

BACKGROUND The Cambridge Cognitive Examination (CAMCOG) is a useful test in screening for Alzheimers disease (AD). However, the interpretation of CAMCOG cut-off scores is problematic and reference values are needed for different educational strata. Given the importance of earlier diagnoses of mild dementia, new cut-off values are required which take into account patients with low levels of education. This study aims to evaluate whether the CAMCOG can be used as an accurate screening test among AD patients and normal controls with different educational levels. METHODS Cross-sectional assessment was undertaken of 113 AD and 208 elderly controls with heterogeneous educational levels (group 1: 1-4 years; group 2: 5-8 years; and group 3: ≥ 9 years) from a geriatric clinic. submitted to a thorough diagnostic evaluation for AD including the Cambridge Examination for Mental Disorders of the Elderly (CAMDEX). Controls had no cognitive or mood complaints. Sensitivity (SE) and specificity (SP) for the CAMCOG in each educational group was assessed with receiver-operator-characteristic (ROC) curves. RESULTS CAMCOG mean values were lower when education was reduced in both diagnostic groups (controls - group 1: 87; group 2: 91; group 3: 96; AD - group 1: 63; group 2: 62; group 3: 77). Cut-off scores for the three education groups were 79, 80 and 90, respectively. SE and SP varied among the groups (group 1: 88.1% and 83.5%; group 2: 84.6% and 96%; group 3: 70.8% and 90%). CONCLUSION The CAMCOG can be used as a cognitive test for patients with low educational level with good accuracy. Patients with higher education showed lower scores than previously reported.

A subtest analysis of the Montreal cognitive assessment (MoCA): which subtests can best discriminate between healthy controls, mild cognitive impairment and Alzheimer's disease?

BACKGROUND It is necessary to continue to explore the psychometric characteristics of key cognitive screening tests such as the Montreal Cognitive Assessment (MoCA) to diagnose cognitive decline as early as possible and to attend to the growing need of clinical trials involving mild cognitive impairment (MCI) participants. The main aim of this study was to assess which MoCA subtests could best discriminate between healthy controls (HC), participants with MCI, and Alzheimers disease (AD). METHODS Cross-sectional analysis of 136 elderly with more than four years of education. All participants were submitted to detailed clinical, laboratory, and neuroimaging evaluation. The MoCA, Mini-Mental State Examination (MMSE), the Cambridge Cognitive Examination (CAMCOG), Geriatric Depression Scale (GDS), and Functional Activities Questionnaire (FAQ) were applied to all participants. The MoCA test was not used in the diagnostic procedure. RESULTS Median MoCA total scores were 27, 23 and 18 for HC, MCI, and AD, respectively (p < 0.001). Word repetition, inverse digits, serial 7, phrases, verbal fluency, abstraction, and word recall discriminated between MCI and HC participants (p < 0.001). The clock drawing, the rhino naming, delayed recall of five words and orientation discriminated between patients with MCI and AD (p < 0.001). A reduced version of the MoCA with only these items did not improve accuracy between MCI and HC (p = 0.076) or MCI and AD (p = 0.119). CONCLUSIONS Not all MoCA subtests might be fundamental to clinical diagnosis of MCI. The reduced versions of MoCA did not add diagnostic accuracy.

Testing allele homogeneity: the problem of nested hypotheses

BackgroundThe evaluation of associations between genotypes and diseases in a case-control framework plays an important role in genetic epidemiology. This paper focuses on the evaluation of the homogeneity of both genotypic and allelic frequencies. The traditional test that is used to check allelic homogeneity is known to be valid only under Hardy-Weinberg equilibrium, a property that may not hold in practice.ResultsWe first describe the flaws of the traditional (chi-squared) tests for both allelic and genotypic homogeneity. Besides the known problem of the allelic procedure, we show that whenever these tests are used, an incoherence may arise: sometimes the genotypic homogeneity hypothesis is not rejected, but the allelic hypothesis is. As we argue, this is logically impossible. Some methods that were recently proposed implicitly rely on the idea that this does not happen. In an attempt to correct this incoherence, we describe an alternative frequentist approach that is appropriate even when Hardy-Weinberg equilibrium does not hold. It is then shown that the problem remains and is intrinsic of frequentist procedures. Finally, we introduce the Full Bayesian Significance Test to test both hypotheses and prove that the incoherence cannot happen with these new tests. To illustrate this, all five tests are applied to real and simulated datasets. Using the celebrated power analysis, we show that the Bayesian method is comparable to the frequentist one and has the advantage of being coherent.ConclusionsContrary to more traditional approaches, the Full Bayesian Significance Test for association studies provides a simple, coherent and powerful tool for detecting associations.

A spectral series approach to high-dimensional nonparametric regression

A key question in modern statistics is how to make fast and reliable inferences for complex, high-dimensional data. While there has been much interest in sparse techniques, current methods do not generalize well to data with nonlinear structure. In this work, we present an orthogonal series estimator for predictors that are complex aggregate objects, such as natural images, galaxy spectra, trajectories, and movies. Our series approach ties together ideas from kernel machine learning, and Fourier methods. We expand the unknown regression on the data in terms of the eigenfunctions of a kernel-based operator, and we take advantage of orthogonality of the basis with respect to the underlying data distribution, P, to speed up computations and tuning of parameters. If the kernel is appropriately chosen, then the eigenfunctions adapt to the intrinsic geometry and dimension of the data. We provide theoretical guarantees for a radial kernel with varying bandwidth, and we relate smoothness of the regression function with respect to P to sparsity in the eigenbasis. Finally, using simulated and real-world data, we systematically compare the performance of the spectral series approach with classical kernel smoothing, k-nearest neighbors regression, kernel ridge regression, and state-of-the-art manifold and local regression methods.

Nonparametric Conditional Density Estimation in a High-Dimensional Regression Setting

In some applications (e.g., in cosmology and economics), the regression is not adequate to represent the association between a predictor x and a response Z because of multi-modality and asymmetry of f(z|x); using the full density instead of a single-point estimate can then lead to less bias in subsequent analysis. As of now, there are no effective ways of estimating f(z|x) when x represents high-dimensional, complex data. In this article, we propose a new nonparametric estimator of f(z|x) that adapts to sparse (low-dimensional) structure in x. By directly expanding f(z|x) in the eigenfunctions of a kernel-based operator, we avoid tensor products in high dimensions as well as ratios of estimated densities. Our basis functions are orthogonal with respect to the underlying data distribution, allowing fast implementation and tuning of parameters. We derive rates of convergence and show that the method adapts to the intrinsic dimension of the data. We also demonstrate the effectiveness of the series method on images, spectra, and an application to photometric redshift estimation of galaxies. Supplementary materials for this article are available online.

A Bayesian Decision-Theoretic Approach to Logically-Consistent Hypothesis Testing

This work addresses an important issue regarding the performance of simultaneous test procedures: the construction of multiple tests that at the same time are optimal from a statistical perspective and that also yield logically-consistent results that are easy to communicate to practitioners of statistical methods. For instance, if hypothesis A implies hypothesis B, is it possible to create optimal testing procedures that reject A whenever they reject B? Unfortunately, several standard testing procedures fail in having such logical consistency. Although this has been deeply investigated under a frequentist perspective, the literature lacks analyses under a Bayesian paradigm. In this work, we contribute to the discussion by investigating three rational relationships under a Bayesian decision-theoretic standpoint: coherence, invertibility and union consonance. We characterize and illustrate through simple examples optimal Bayes tests that fulfill each of these requisites separately. We also explore how far one can go by putting these requirements together. We show that although fairly intuitive tests satisfy both coherence and invertibility, no Bayesian testing scheme meets the desiderata as a whole, strengthening the understanding that logical consistency cannot be combined with statistical optimality in general. Finally, we associate Bayesian hypothesis testing with Bayes point estimation procedures. We prove the performance of logically-consistent hypothesis testing by means of a Bayes point estimator to be optimal only under very restrictive conditions.

Learning with many experts: Model selection and sparsity

Experts classifying data are often imprecise. Recently, several models have been proposed to train classifiers using the noisy labels generated by these experts. How to choose between these models? In such situations, the true labels are unavailable. Thus, one cannot perform model selection using the standard versions of methods such as empirical risk minimization and cross validation. In order to allow model selection, we present a surrogate loss and provide theoretical guarantees that assure its consistency. Next, we discuss how this loss can be used to tune a penalization which introduces sparsity in the parameters of a traditional class of models. Sparsity provides more parsimonious models and can avoid overfitting. Nevertheless, it has seldom been discussed in the context of noisy labels due to the difficulty in model selection and, therefore, in choosing tuning parameters. We apply these techniques to several sets of simulated and real data.