Juan Carlos Correa

Random Forests to Predict Rectal Toxicity Following Prostate Cancer Radiation Therapy

PURPOSE To propose a random forest normal tissue complication probability (RF-NTCP) model to predict late rectal toxicity following prostate cancer radiation therapy, and to compare its performance to that of classic NTCP models. METHODS AND MATERIALS Clinical data and dose-volume histograms (DVH) were collected from 261 patients who received 3-dimensional conformal radiation therapy for prostate cancer with at least 5 years of follow-up. The series was split 1000 times into training and validation cohorts. A RF was trained to predict the risk of 5-year overall rectal toxicity and bleeding. Parameters of the Lyman-Kutcher-Burman (LKB) model were identified and a logistic regression model was fit. The performance of all the models was assessed by computing the area under the receiving operating characteristic curve (AUC). RESULTS The 5-year grade ≥2 overall rectal toxicity and grade ≥1 and grade ≥2 rectal bleeding rates were 16%, 25%, and 10%, respectively. Predictive capabilities were obtained using the RF-NTCP model for all 3 toxicity endpoints, including both the training and validation cohorts. The age and use of anticoagulants were found to be predictors of rectal bleeding. The AUC for RF-NTCP ranged from 0.66 to 0.76, depending on the toxicity endpoint. The AUC values for the LKB-NTCP were statistically significantly inferior, ranging from 0.62 to 0.69. CONCLUSIONS The RF-NTCP model may be a useful new tool in predicting late rectal toxicity, including variables other than DVH, and thus appears as a strong competitor to classic NTCP models.

Pooling/bootstrap-based GWAS (pbGWAS) identifies new loci modifying the age of onset in PSEN1 p.Glu280Ala Alzheimer's disease

The literature on GWAS (genome-wide association studies) data suggests that very large sample sizes (for example, 50,000 cases and 50,000 controls) may be required to detect significant associations of genomic regions for complex disorders such as Alzheimers disease (AD). Because of the challenges of obtaining such large cohorts, we describe here a novel sequential strategy that combines pooling of DNA and bootstrapping (pbGWAS) in order to significantly increase the statistical power and exponentially reduce expenses. We applied this method to a very homogeneous sample of patients belonging to a unique and clinically well-characterized multigenerational pedigree with one of the most severe forms of early onset AD, carrying the PSEN1 p.Glu280Ala mutation (often referred to as E280A mutation), which originated as a consequence of a founder effect. In this cohort, we identified novel loci genome-wide significantly associated as modifiers of the age of onset of AD (CD44, rs187116, P=1.29 × 10−12; NPHP1, rs10173717, P=1.74 × 10−12; CADPS2, rs3757536, P=1.54 × 10−10; GREM2, rs12129547, P=1.69 × 10−13, among others) as well as other loci known to be associated with AD. Regions identified by pbGWAS were confirmed by subsequent individual genotyping. The pbGWAS methodology and the genes it targeted could provide important insights in determining the genetic causes of AD and other complex conditions.

Placing joy, surprise and sadness in space: a cross-linguistic study.

The valence–space metaphor posits that emotion concepts map onto vertical space such that positive concepts are in upper locations and negative in lower locations. Whilst previous studies have demonstrated this pattern for positive and negative emotions e.g. ‘joy’ and ‘sadness’, the spatial location of neutral emotions, e.g. ‘surprise’, has not been investigated, and little is known about the effect of linguistic background. In this study, we first characterised the emotions joy, surprise and sadness via ratings of their concreteness, imageability, context availability and valence before examining the allocation of these emotions in vertical space. Participants from six linguistic groups completed either a rating task used to characterise the emotions or a word allocation task to implicitly assess where these emotions are positioned in vertical space. Our findings suggest that, across languages, gender, handedness, and ages, positive emotions are located in upper spatial locations and negative emotions in lower spatial locations. In addition, we found that the neutral emotional valence of surprise is reflected in this emotion being mapped mid-way between upper and lower locations onto the vertical plane. This novel finding indicates that the location of a concept on the vertical plane mimics the concept’s degree of emotional valence.

Spatial nonparametric mixed-effects model with spatial-varying coefficients for analysis of population

Voxel-wise comparisons have been largely used in the analysis of populations to identify biomarkers for pathologies, disease progression, or to predict a treatment outcome. On the basis of a good interindividual spatial alignment, 3D maps are produced, allowing to localise regions where significant differences between groups exist. However, these techniques have received some criticism as they rely on conditions wich are not always met. Firstly, the results may be affected by misregistrations; secondly, the statistics behind the models assumes normally distributed data; finally, because of the size of the images, some strategies must be used to control for the rate of false detection. In this paper, we propose a spatial (3D) nonparametric mixed-effects model for population analysis. It overcomes some of the issues of classical voxel-based approaches, namely robustness to false positive rates, misregistrations and large variances between groups. Examples on numerical phantoms and real clinical data illustrate the feasiblity of the approach. An example of application within the development of voxel-wise predictive models of rectal toxicity in prostate cancer radiotherapy is presented. Results demonstrate an improved sensitivity and reliability for group analysis compared with standard voxel-wise methods and open the way for potential applications in computational anatomy.

A tensor-based population value decomposition to explain rectal toxicity after prostate cancer radiotherapy.

In prostate cancer radiotherapy the association between the dose distribution and the occurrence of undesirable side-effects is yet to be revealed. In this work a method to perform population analysis by comparing the dose distributions is proposed. The method is a tensor-based approach that generalises an existing method for 2D images and allows for the highlighting of over irradiated zones correlated with rectal bleeding after prostate cancer radiotherapy. Thus, the aim is to contribute to the elucidation of the dose patterns correlated with rectal toxicity. The method was applied to a cohort of 63 patients and it was able to build up a dose pattern characterizing the difference between patients presenting rectal bleeding after prostate cancer radiotherapy and those who did not.

Manipulating the alpha level cannot cure significance testing

We argue that making accept/reject decisions on scientific hypotheses, including a recent call for changing the canonical alpha level from p = 0.05 to p = 0.005, is deleterious for the finding of new discoveries and the progress of science. Given that blanket and variable alpha levels both are problematic, it is sensible to dispense with significance testing altogether. There are alternatives that address study design and sample size much more directly than significance testing does; but none of the statistical tools should be taken as the new magic method giving clear-cut mechanical answers. Inference should not be based on single studies at all, but on cumulative evidence from multiple independent studies. When evaluating the strength of the evidence, we should consider, for example, auxiliary assumptions, the strength of the experimental design, and implications for applications. To boil all this down to a binary decision based on a p-value threshold of 0.05, 0.01, 0.005, or anything else, is not acceptable.

A new approach to the Box–Cox transformation

We propose a new methodology to estimate λ, the parameter of the Box-Cox transformation, as well as an alternative method to determine plausible values for it. The former is accomplished by defining a grid of values for λ and further perform a normality test on the λ -transformed data. The optimum value of λ, say λ * , is such that the p-value from the normality test is the highest. The set of plausible values is determined using the inverse probability method after plotting the p-values against the values of λ on the grid. Our methodology is illustrated with two real-world data sets. Furthermore, a simulation study suggests that our method improves the symmetry, kurtosis and, hence, the normality of data, making it a feasible alternative to the traditional Box-Cox transformation.

Información relevante para la gestión por calidad en las organizaciones de salud (Relevant information management for quality health organizations) (Gestão de informações relevantes para organizações de saúde de qualidade)

Objetivo. Someter al escrutinio de la comunidad cientifica una matriz construida con la informacion de interes en la gestion por calidad de las organizaciones de salud, como recurso de homologacion y estandarizacion de unos indices cuya definicion, bondad de ajuste e interrelacion carecen de antecedentes en la literatura especializada. Materiales y metodos. 1. Revision y sustentacion de la informacion considerada relevante para el efecto. 2. Diseno y presentacion de la matriz con los seis (6) indices seleccionados: Oportunidad, Riesgo, Gestion, Satisfaccion, Innovacion y Ambiental, para un total de 14 indicadores, relativos a estructura, procesos y resultados dentro de las tres areas clave de una organizacion de salud tipo: Asistencial, Estrategico/Administrativa y Laboral. Resultados: Matriz base del Sistema de Indices de Gestion por Calidad para las Organizaciones de Salud (SIGNOS), actualmente a prueba y proximo a ser publicado. Conclusiones. Si bien los indices establecidos no son comunmente manejados, por su pertinencia, coherencia y, en especial, por su probada funcionalidad dentro del Sistema antedicho, se constituyen en una valiosa herramienta para la gestion por calidad, lo mismo que para la clasificacion de las empresas de salud. Abstract Aim. Subject to scrutiny by the scientific community constructed a matrix with information of interest in quality management by health organizations as a resource of certification and standardization of the definition indices, goodness of fit and interplay lack background in literature. Materials and methods. 1. Review and support of information considered relevant to the effect. 2. Design and layout of the array with six (6) selected indices: Opportunity, Risk, Management, Satisfaction, Innovation and Environment, for a total of 14 indicators relating to structure, processes and outcomes within three key areas of an organization health type: Welfare, Strategic / Management and Labor. Results. Based Matrix Indicator System for Quality Management for Healthcare Organizations (ISQMHO), currently being tested and soon to be published. Conclusions. Although the rates established are not commonly managed by relevance, consistency and, in particular, for its proven functionality within the system above, constitute a valuable tool for quality management, as well as for classifying company’s health. Resumo Objetivo. Sujeita ao escrutinio da comunidade cientifica construida uma matriz com informacoes de interesse na gestao da qualidade pelas organizacoes de saude como um recurso de certificacao e padronizacao dos indices de definicao, a bondade de ajuste e de funda interacao falta na literatura. Materiais e metodos. 1. Revisao e apoio de informacoes consideradas relevantes para o efeito. 2. Design e layout da matriz com seis (6) os indices selecionados: Oportunidade, Risco, Gestao, Satisfacao, Inovacao e Meio Ambiente, para um total de 14 indicadores relativos a estrutura, processos e resultados em tres areas-chave de uma organizacao tipo de saude: Bem-estar, Gestao / Estrategica e Trabalho. Resultados. Sistema de Indicadores baseado Matriz de Gestao da Qualidade para Organizacoes de Saude (MGQOS), atualmente sendo testado e que em breve sera publicado. Conclusoes. Embora as taxas estabelecidas nao sejam comumente geridas por consistencia relevância, e, em particular, para a sua funcionalidade comprovada no interior do sistema acima, constitui uma ferramenta valiosa para a gestao da qualidade, bem como para a classificacao de empresas saude. Palavras Chave: Gerencia, Gestao de Qualidade, Gestao em Saude, Gerenciamento de Informacao, Indicadores de Gestao

Cross validation of pooling/resampling GWAS using the WTCCC data

Recently, we presented a new method of pooling/resampling genome-wide association study (prGWAS) that uncovered new and known loci associated to Alzheimer’s disease. In here, we contrast this method with the Welcome Trust Case Control Consortium (WTCCC) data, a well-known GWAS on seven human complex diseases. Our results suggest that prGWAS can be considered an efficient, specific, and accurate alternative to the conventional GWAS approach at a fraction of the genotyping cost, and provide insights into other potential applications such as next generation sequencing.

A non parametric mixed-effect model for population analysis: Application to Alzheimer's disease data

In population analysis, the images of different groups can be compared to locate the effects of a particular disease or treatment and also to generate biomarkers that help in the diagnosis process. Voxel-Based Morphometry (VBM) is a set of widely extended techniques to compare groups of images. VBM involves image normalization, image smoothing, statistical map generation and correction for hypothesis testing. In this paper, we propose the use of a nonparametric mixed-effect model to study Alzheimers Disease (AD). The proposed method can handle covariates and through the integration of the smoothing and statistical map generation, individual specificities can be controlled. Moreover, it allows the reconstruction of the typical shapes for each group and it can be advantageously used as another VBM implementation.