Tesi sul tema "Données de survies"
Cita una fonte nei formati APA, MLA, Chicago, Harvard e in molti altri stili
Vedi i top-50 saggi (tesi di laurea o di dottorato) per l'attività di ricerca sul tema "Données de survies".
Accanto a ogni fonte nell'elenco di riferimenti c'è un pulsante "Aggiungi alla bibliografia". Premilo e genereremo automaticamente la citazione bibliografica dell'opera scelta nello stile citazionale di cui hai bisogno: APA, MLA, Harvard, Chicago, Vancouver ecc.
Puoi anche scaricare il testo completo della pubblicazione scientifica nel formato .pdf e leggere online l'abstract (il sommario) dell'opera se è presente nei metadati.
Vedi le tesi di molte aree scientifiche e compila una bibliografia corretta.
Ortholand, Juliette. "Joint modelling of events and repeated observations : an application to the progression of Amyotrophic Lateral Sclerosis". Electronic Thesis or Diss., Sorbonne université, 2024. http://www.theses.fr/2024SORUS227.
Testo completoProgression heterogeneity in chronic diseases such as Amyotrophic Lateral Sclerosis (ALS) is a significant obstacle to developing effective treatments. Leveraging the growing wealth of large databases through modelling can help better understanding it. However, the data collected only offer access to partial trajectories, that need to be realigned to reconstruct a comprehensive disease progression. To address this challenge, data-driven progression models like the longitudinal Spatiotemporal model were developed. Its main interest is its ability to synchronise patients onto a common disease timeline (temporal aspect) thanks to a latent disease age, while also capturing the remaining variability through parameters that account for outcome ordering (spatial aspect). However, this model was primarily designed for longitudinal data, overlooking crucial outcomes in ALS such as time to death or initiation of life support, like Non-Invasive Ventilation (NIV). Conversely, existing joint models offer the advantage of simultaneously handling longitudinal and survival data. However, they do not realign trajectories, which compromises their temporal resolution. This thesis aimed to expand the Spatiotemporal model into a Joint Spatiotemporal model, enabling, for ALS research, the examination of survival data alongside longitudinal data. First, we applied the Spatiotemporal model to explore how the interaction between sex and onset site (spinal or bulbar) impacts the progression of ALS patients. We selected 1,438 patients from the PRO-ACT database. We demonstrated a significant influence of both sex and onset site on six longitudinal outcomes monitoring the functional and respiratory decline in addition to Body Mass Index. However, this study did not incorporate survival analysis, despite its paramount importance in ALS, due to limitations inherent to the Spatiotemporal model. To address this gap, we associated the Spatiotemporal model with a survival model that estimates a Weibull survival model from its latent disease age, creating a univariate Joint Temporal model. After model validation, we benchmarked our model with a state-of-the-art joint model on PRO-ACT data. Our model exhibited significantly superior performance in terms of absolute bias and mean cumulative AUC for right-censored events. This demonstrated the efficacy of our approach in the context of ALS compared to existing joint models. However, modelling several longitudinal outcomes requires a multivariate approach. Life support initiation that might be censored by death needs to be also considered. We thus extended the Joint Temporal model, into a multivariate Joint Spatiotemporal model with competing risks to analyse NIV initiation. This involved coupling the multivariate Spatiotemporal model with a cause-specific Weibull survival model from the latent disease age. We incorporated spatial parameters with a Cox proportional effect on the hazard. After validation, we benchmarked our model with a state-of-the-art joint model on PRO-ACT data and analysed sex and onset site interaction in complement to the first study. The Joint Spatiotemporal model achieved similar performance to the state-of-the-art model while capturing an underlying shared latent process, the latent disease age, whereas the state-of-the-art models the impact of longitudinal outcomes on survival. To enhance the reproducibility and facilitate the reuse of these models, the proposed models were implemented in the open-source software Leaspy. In conclusion, this thesis introduces the first data-driven progression model combining longitudinal and survival modelling. We demonstrated its relevance to understand the occurrence of critical events in ALS. This work paves the way for further extension to analyse recurrent events, among other potential applications in causal inference
Martin, Marie-Laure. "Données de survie". Paris 11, 2001. http://www.theses.fr/2001PA112335.
Testo completoWe consider two statistical problems arising during the estimation of the hazard function of cancer death in Hiroshima. The first problem is the estimation of the hazard function when the covariate is mismeasured. In Chapter 2, only grouped data are available, and the mismeasurement of the covariate is modeled as a misclassification. An easily implemented estimation procedure based on a generalization of the least squares method is devised for estimating simultaneously the parameters of the hazard function and the misclassification probabilities. The procedure is applied for taking into account the mismeasurement of the dose of radiation in the estimation of the hazard function of solid cancer death in Hiroshima. In Chapter 3 available data are individual data. We consider a model of excess relative risk, and we assume that the covariate is measured with a Gaussian additive error. We propose an estimation criterion based on the partial log-likelihood, and we show that the estimator obtained by maximization of this criterion is consistent and asymptotically Gaussian. Our result extends to other polynomial regression functions, to the Cox model and to the log-normal error model. The second problem is the non-parametric estimation of the hazard function. We consider the model of excess relative and absolute risk and propose a non-parametric estimation of the effect of the covariate using a model selection procedure, when available data are stratified data. We approximate the function of the covariate by a collection of spline functions, and select the best one according to Akaike Information Criterion. By the same way we choose which model between the model of excess relative risk or excess absolute risk fits the best the data. We apply our method for estimating the solid cancer and leukemia death hazard functions in Hiroshima
Heutte, Natacha. "Modèles semi-markoviens et données de survie". Paris 5, 2001. http://www.theses.fr/2001PA05S013.
Testo completoModels for survival data including a categorized quality of life index is proposed. The model is intended to take into account the effect of endogenous and exogenous factors both on the duration of survival and the quality of life. Endogenous factors are for example biological measurements or genetical specifications, while exogenous ones are environmental factors. The proposed models are semi-parametric and based on semi-markov processes. Time may be continuous or discrete depending on the type of the data. The general framework of all preexisting models is sketched. Estimators are derived, as well as their asymptotic properties, and algorithms and programs are given to compute them explicitly. They are exemplified on real data on AIDS and cancer patients, and on simulations. Those models are presented in a biomedical context but can be useful in any field where durations together with multistate processes are involved
Lorino, Tristan. "Modèles statistiques pour des données de survie corrélées". Phd thesis, Institut national agronomique paris-grignon - INA P-G, 2002. http://tel.archives-ouvertes.fr/tel-00003672.
Testo completoNous étudions les deux principales classes de modèles pour données de survie corrélées : les modèles de fragilité (ou conditionnels) et les modèles marginaux. Nous nous proposons une large comparaison de ces deux approches, d'une part au travers d'une étude de données vétérinaires, d'autre part au moyen de simulations.
Notre objectif est d'évaluer la sensibilité de tels modèles vis-à-vis de la structure des jeux de données qu'ils sont appelés à traiter -- et plus particulièrement vis-à-vis de la taille des groupes.
Dantan, Etienne. "Modèles conjoints pour données longitudinales et données de survie incomplètes appliqués à l'étude du vieillissement cognitif". Thesis, Bordeaux 2, 2009. http://www.theses.fr/2009BOR21658/document.
Testo completoIn cognitive ageing study, older people are highly selected by a risk of death associated with poor cognitive performances. Modeling the natural history of cognitive decline is difficult in presence of incomplete longitudinal and survival data. Moreover, the non observed cognitive decline acceleration beginning before the dementia diagnosis is difficult to evaluate. Cognitive decline is highly heterogeneous, e.g. there are various patterns associated with different risks of survival event. The objective is to study joint models for incomplete longitudinal and survival data to describe the cognitive evolution in older people. Latent variable approaches were used to take into account the non-observed mechanisms, e.g. heterogeneity and decline acceleration. First, we compared two approaches to consider missing data in longitudinal data analysis. Second, we propose a joint model with a latent state to model cognitive evolution and its pre-dementia acceleration, dementia risk and death risk
Bazin, Gurvan. "Analyse différée des données du SuperNova Legacy Survey". Paris 7, 2008. http://www.theses.fr/2008PA077135.
Testo completoThe SuperNova Legacy Survey (SNLS) experiment observed type la supemovae (SNeHa) during 5 years. Its aim is the contraint cosmological parameters. The online reduction pipeline is based on spectroscopic identification for each supernova. Systematically using spectroscopy requires a sufficient signal to noise level. Thus, it could lead to selection biases and would not be possible for future surveys The PhD thesis report a complementary method for data reduction based on a completely photometric selection. This analysis, more efficient to select faint events, approximately double the SNeHa sample of the SNLS. This method show a clear bias in the spectroscopic selection. Brighter SNeHa are systematically selected beyond a redshift of 0. 7. On the other hand, no important impact on cosmology was found. So, corrections on intrinsic variability of SNeHa luminosity are robust. In addition, this work is a first step to study the feasibility of such a purely photometric analysis for cosmology. This is a promising method for future projects
Bach, Aurélien. "Prédiction de survie sur des données cliniques censurées et application à la MPOC". Mémoire, Université de Sherbrooke, 2017. http://hdl.handle.net/11143/11580.
Testo completoStamenic, Danko. "Modélisation conjointe pour données longitudinales et données de survie : analyse des facteurs prédictifs du devenir de la greffe rénale". Thesis, Limoges, 2018. http://www.theses.fr/2018LIMO0027/document.
Testo completoPrediction of graft outcome would be useful to optimize patient care. Follow-up of kidneytransplant patients include repeated measurements of longitudinal markers, such as serum creatinine and immunosuppressive drug exposure. Recently proposed joint models areappropriate to analyze relationship between longitudinal processes and time-to-event data. In the first part of present work, we used the approach of joint latent class mixed models tostudy the impact of time-profiles of serum creatinine collected within the first 18 months after kidney transplantation on long-term graft survival. The studied cohort was parted into three homogenous classes with a specific time-evolution of serum creatinine and a specific risk of graft failure. The individual predicted probabilities of graft failure up to 10 years posttransplantation, calculated from this joint model were satisfying in terms of sensitivity, specificity and overall accuracy, for patients who had not developed de novo donor specificanti-HLA antibodies. The clinical usefulness of developed predictive tooI needs to beevaluated with a dynamic approach. In the second part, non-linear mixed effects models witha mixture of distribution for random effects were used to investigate (i) the associationbetween variability over time of tacrolimus exposure and self-reported drug adherence and(ii) the impact of this variability on the acute rejection risk. This model found a significantimpact of tacrolimus time-exposure variability on acute rejection onset beyond 3 months posttransplantation. On the contrary, no association between adherence and (i) variability oftacrolimus time-exposure and (ii) acute rejection was observed in our study which included moderate non-adherent patients only. This result questions the impact of moderate nonadherence on graft outcome
Desmée, Solène. "Modélisation conjointe de données longitudinales non-linéaires et de données de survie : application au cancer de la prostate métastatique". Sorbonne Paris Cité, 2016. http://www.theses.fr/2016USPCC115.
Testo completoTreatment evaluation for metastatic Castration-Resistant Prostate Cancer (mCRPC) relies on time-to-death. Prostate-specific antigen (PSA), assumed to be linked to survival, is frequently measured. Joint modelling which consists in the simultaneous analyse of biomarker's evolution and survival is particularly adapted, but often limited to linear longitudinal process. The objective of this PhD is to study joint modelling when biomarker kinetics is described by a nonlinear mixed-effects model (NLMEM). We established by simulations that the SAEM algorithm of Monolix provided unbiased parameter estimations of a nonlinear joint model, with satisfying type 1 error and power to detect a link between the processes. Then we developed a mechanistic joint model to characterize the relationship between PSA kinetics and survival in mCRPC patients treated by docetaxel. The structural model of the NLMEM was defined by a system of differential equations (DE) describing the mechanism of PSA production by docetaxel-sensitive and -resistant cells. Model selection and evaluation were detailed. The final model showed the predominant role of the non-observed resistant cells on survival. Lastly we expanded tools developed in a linear context for individual dynamic prediction using nonlinear joint model. A Bayesian method provided the distribution of individual parameters. Predictive performances of the model were assessed using time-dependent discrimination and calibration metrics. These works open the way for the development of mechanistic joint models, which enable to account for the impact of several biomarkers on survival through DE, in order to improve therapeutic evaluation and prediction
Lopez, Olivier. "Réduction de dimension en présence de données censurées". Phd thesis, Rennes 1, 2007. http://tel.archives-ouvertes.fr/tel-00195261.
Testo completovariable explicative. Nous développons une nouvelle approche de réduction de la dimension afin de résoudre ce problème.
Katsahian, Sandrine. "Modèles statistiques de survie pour données corrélées : l'exemple de la greffe de moelle". Paris 6, 2007. http://www.theses.fr/2007PA066620.
Testo completoRousseau, Vanessa. "Modèles fragilisés de survie relative appliqués aux données censurées dans le cas de la transplantation rénale". Montpellier 1, 2008. http://www.theses.fr/2008MON1T012.
Testo completoMabika, Bienvenu. "Analyse bayésienne des données de survie : Application à des essais cliniques en pharmacologie". Rouen, 1999. http://www.theses.fr/1999ROUES100.
Testo completoTwo distinct sections constitute this thesis : the first section deals with the Bayesian procedures of survival data and the second with the applying procedures of Bayesian methods. The methods are illustrated with some examples of a mortality study in cardiologic and cancer research where a new treatment is compared to a standard treatment. In the first section, we study the Bayesian framework allowing to compare two Weibull survival distributions with unequal shape parameters, in the case of right censored survival data obtained for two independent samples is considered. For a family of appropriate priors we give the posterior distributions and the highest posterior density intervals about relevant parameters allowing to search for a conclusion of clinical superiority of the treatment. We introduce a Bayesian estimator of the survival function. We propose and study a Bayesian testing for equivalence of two survival functions and an algorithm using Gibbs sampling allowed to resolve this test. Moreover the predictive distributions are used to obtain an early stopping rule in the case of interim analyses. On using some criteria of determination of number patient we propose two approaches. We generalize the Weibull model by a Bayesian approach with covariates, this approach is like to Cox model. The methods inference used there do appeal to Markov Chain Monte Carlo methods
Fourmanoit, N. "Analyse des 5 ans de données de l'expérience SuperNova Legacy Survey". Phd thesis, Université Pierre et Marie Curie - Paris VI, 2010. http://tel.archives-ouvertes.fr/tel-00587450.
Testo completoFourmanoit, Nicolas. "Analyse des 5 ans de données de l'expérience SuperNova Legacy Survey". Paris 6, 2010. http://www.theses.fr/2010PA066282.
Testo completoLeconte-Lavaud, Evelyne. "Modeles et tests pour l'analyse statistique des donnees censurees multivariees". Paris 11, 1995. http://www.theses.fr/1995PA11T014.
Testo completoRouam, Sigrid Laure. "Développement d'un indice de séparabilité adapté aux données de génomique en analyse de survie". Phd thesis, Université Paris Sud - Paris XI, 2011. http://tel.archives-ouvertes.fr/tel-00718743.
Testo completoChagny, Gaëlle. "Estimation adaptative avec des données transformées ou incomplètes. Application à des modèles de survie". Phd thesis, Université René Descartes - Paris V, 2013. http://tel.archives-ouvertes.fr/tel-00863141.
Testo completoRouam, Sigrid Laure. "Développement d’un indice de séparabilité adapté aux données de génomique en analyse de survie". Thesis, Paris 11, 2011. http://www.theses.fr/2011PA11T006/document.
Testo completoIn oncogenomics research, one of the main objectives is to identify new genomic markers so as to construct predictive rules in order to classify patients according to time-to-event outcomes (death or tumor relapse). Most of the studies dealing with such high throughput data usually rely on a selection process in order to identify, among the candidates, the markers having a prognostic impact. A common problem among biologists is the choice of the selection rule. In survival analysis, classical procedures consist in ranking genetic markers according to either the estimated hazards ratio or quantities derived from a test statistic (p-value, q-value). However, these methods are not suitable for gene selection across multiple genomic datasets with different sample sizes.Using an index taking into account the magnitude of the prognostic impact of factors without being highly dependent on the sample size allows to address this issue. In this work, we propose a novel index of predictive ability for selecting genomic markers having a potential impact on timeto-event outcomes. This index extends the notion of "pseudo-R2" in the ramework of survival analysis. It possesses an original and straightforward interpretation in terms of "separability". The index is first derived in the framework of the Cox model and then extended to more complex non-proportional hazards models. Simulations show that our index is not substantially affected by the sample size of the study and the censoring. They also show that its separability performance is higher than indices from the literature. The interest of the index is illustrated in two examples. The first one aims at identifying genomic markers with common effects across different cancertypes. The second shows, in the framework of a lung cancer study, the interest of the index for selecting genomic factor with crossing hazards functions, which could be explained by some "modulating" effects between markers. The proposed index is a promising tool, which can help researchers to select a list of features of interest for further biological investigations
Belot, Aurélien. "Modélisation flexible des données de survie en présence de risques concurrents et apports de la méthode du taux en excès". Aix-Marseille 2, 2009. http://www.theses.fr/2009AIX20709.
Testo completoIn epidemiology, the probability of survival (associated to the delay until death) of a cohort of patients is a key indicator of the impact of the disease. But, this survival may be estimated according to various causes of death; these constitute then competing events. In this dissertation, after presenting the analysis setting, we propose a flexible model to estimate jointly the hazards of competing events as well as the effects of prognostic factors in function of the time elapsed since diagnosis. Furthermore, this model allows comparing the effects of the prognostic factors on the competing events; it was applied to an analysis of data on an American cohort of patients with colorectal cancer. However, the causes of death may sometimes be missing or invalid (case of registries that do not routinely collect the causes of death). The statistical method of the excess hazard makes it possible to overcome the need for the causes of death by using the general population mortality to estimate the excess mortality directly or indirectly linked to the disease. An analysis strategy is proposed to estimate the excess mortality as well as the non-linear and/or time-dependent effects of the prognostic factors. In addition to death, the competing events method is also applied to intercurrent events such as relapse or metastasis. A model that combines the competing events and the excess hazard methods is proposed to estimate the hazards of intercurrent events and the excess mortality; it is applied to data from FRANCIM registries on colorectal cancer cases with curative-intent treatment
Héraud, Bousquet Vanina. "Traitement des données manquantes en épidémiologie : application de l’imputation multiple à des données de surveillance et d’enquêtes". Thesis, Paris 11, 2012. http://www.theses.fr/2012PA11T017/document.
Testo completoThe management of missing values is a common and widespread problem in epidemiology. The most common technique used restricts the data analysis to subjects with complete information on variables of interest, which can reducesubstantially statistical power and precision and may also result in biased estimates.This thesis investigates the application of multiple imputation methods to manage missing values in epidemiological studies and surveillance systems for infectious diseases. Study designs to which multiple imputation was applied were diverse: a risk analysis of HIV transmission through blood transfusion, a case-control study on risk factors for ampylobacter infection, and a capture-recapture study to estimate the number of new HIV diagnoses among children. We then performed multiple imputation analysis on data of a surveillance system for chronic hepatitis C (HCV) to assess risk factors of severe liver disease among HCV infected patients who reported drug use. Within this study on HCV, we proposedguidelines to apply a sensitivity analysis in order to test the multiple imputation underlying hypotheses. Finally, we describe how we elaborated and applied an ongoing multiple imputation process of the French national HIV surveillance database, evaluated and attempted to validate multiple imputation procedures.Based on these practical applications, we worked out a strategy to handle missing data in surveillance data base, including the thorough examination of the incomplete database, the building of the imputation model, and the procedure to validate imputation models and examine underlying multiple imputation hypotheses
Allart, Thibault. "Apprentissage statistique sur données longitudinales de grande taille et applications au design des jeux vidéo". Electronic Thesis or Diss., Paris, CNAM, 2017. http://www.theses.fr/2017CNAM1136.
Testo completoThis thesis focuses on longitudinal time to event data possibly large along the following tree axes : number of individuals, observation frequency and number of covariates. We introduce a penalised estimator based on Cox complete likelihood with data driven weights. We introduce proximal optimization algorithms to efficiently fit models coefficients. We have implemented thoses methods in C++ and in the R package coxtv to allow everyone to analyse data sets bigger than RAM; using data streaming and online learning algorithms such that proximal stochastic gradient descent with adaptive learning rates. We illustrate performances on simulations and benchmark with existing models. Finally, we investigate the issue of video game design. We show that using our model on large datasets available in video game industry allows us to bring to light ways of improving the design of studied games. First we have a look at low level covariates, such as equipment choices through time and show that this model allows us to quantify the effect of each game elements, giving to designers ways to improve the game design. Finally, we show that the model can be used to extract more general design recommendations such as dificulty influence on player motivations
Allart, Thibault. "Apprentissage statistique sur données longitudinales de grande taille et applications au design des jeux vidéo". Thesis, Paris, CNAM, 2017. http://www.theses.fr/2017CNAM1136/document.
Testo completoThis thesis focuses on longitudinal time to event data possibly large along the following tree axes : number of individuals, observation frequency and number of covariates. We introduce a penalised estimator based on Cox complete likelihood with data driven weights. We introduce proximal optimization algorithms to efficiently fit models coefficients. We have implemented thoses methods in C++ and in the R package coxtv to allow everyone to analyse data sets bigger than RAM; using data streaming and online learning algorithms such that proximal stochastic gradient descent with adaptive learning rates. We illustrate performances on simulations and benchmark with existing models. Finally, we investigate the issue of video game design. We show that using our model on large datasets available in video game industry allows us to bring to light ways of improving the design of studied games. First we have a look at low level covariates, such as equipment choices through time and show that this model allows us to quantify the effect of each game elements, giving to designers ways to improve the game design. Finally, we show that the model can be used to extract more general design recommendations such as dificulty influence on player motivations
Rondeau, Virginie. "Analyse par vraisemblance pénalisée de données de survie groupées : application à la relation entre aluminium et démence". Bordeaux 2, 2000. http://www.theses.fr/2000BOR28719.
Testo completoLueza, Béranger. "Estimation du bénéfice de survie à partir de méta-analyse sur données individuelles et évaluation économique". Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLS268/document.
Testo completoThe survival benefit restricted up to a certain time horizon has been suggested as an alternative measure to the common relative measures used to estimate the treatment effect, especially in case of non-proportional hazards of death. The restricted survival benefit corresponds to the difference of the two restricted mean survival times estimated for each treatment arm, and is expressed in terms of life years gained. In the literature, this measure is considered as more intuitive than the hazard ratio and many authors have suggested its use for the design and the analysis of clinical trials. However, it is not currently the most used measure in randomized trials. This measure is valid under any distribution of the survival times and is adapted if the proportional hazards assumption does not hold. In addition, the restricted survival benefit can be used in medico-economic evaluation where an absolute measure of the treatment effect is needed (number of [quality adjusted] life years gained). If one wants to estimate the restricted survival benefit from an individual participant data meta-analysis, there is a need to take into account the trial effect due to the hierarchical structure of the data. The aim of this thesis was to compare statistical methods to estimate the restricted survival benefit from an individual participant data meta-analysis of randomized trials. The starting point was a case study (cost-effectiveness analysis) using data from the Meta-Analysis of Radiotherapy in Lung Cancer. This study showed that the five investigated methods yielded different estimates for the restricted survival benefit and its confidence interval. The choice of a method to estimate the survival benefit also impacted on cost-effectiveness results. Our second project consisted in a simulation study to have a better understanding of the properties of the investigated methods in terms of bias and standard error. Finally, the last part of the thesis illustrated the lessons learned from the simulation study through three examples of individual participant data meta-analysis in nasopharynx cancer and in small cell lung cancer
Roué, Tristan. "Épidémiologie des cancers en Guyane : Analyse des données du registre des cancers de Guyane". Thesis, Antilles-Guyane, 2014. http://www.theses.fr/2014AGUY0743/document.
Testo completoThe objective of the cancer registry of French Guiana is to compile all patients living in French Guiana with malignant invasive pathology and/or in situ lesions starting January 1st 2003 in persons living in French Guiana, whatever the tumoral location and the place of diagnosis and care. This study aimed to describe the population with invasive cancer to improve the knowledge about this disease in order to target public health interventions more effectively.The age standardised incidence rate was 30% times lower than in France in both sexes and the same than in South America.We compared incidence and relative survival of patients with invasive breast cancer (IBC) and patients with invasive cervical cancer (ICC) between women from French Guiana and metropolitan France.The ratio between incidence and mortality showed that the prognosis of IBC in French Guiana was worse than in metropolitan France.The relative survival rate among women with IBC in French Guiana was lower than among women in metropolitan France.In French Guiana, the age-standardized incidence rate of cervical cancer was four times higher than in France. Women living in remote areas seemed to be diagnosed later and more often following symptoms.Access to care for migrants is challenging and sustains health inequalities. Early detection through prevention programs is crucial for increasing cancer survival notably for foreign-born patients. Further studies with more patients and other variables could improve the knowledge about these diseases
Jaunâtre, Kévin. "Analyse et modélisation statistique de données de consommation électrique". Thesis, Lorient, 2019. http://www.theses.fr/2019LORIS520.
Testo completoIn October 2014, the French Environment & Energy Management Agency with the ENEDIS company started a research project named SOLENN ("SOLidarité ENergie iNovation") with multiple objectives such as the study of the control of the electric consumption by following the households and to secure the electric supply. The SOLENN project was lead by the ADEME and took place in Lorient, France. The main goal of this project is to improve the knowledge of the households concerning the saving of electric energy. In this context, we describe a method to estimate extreme quantiles and probabilites of rare events which is implemented in a R package. Then, we propose an extension of the famous Cox's proportional hazards model which allows the etimation of the probabilites of rare events. Finally, we give an application of some statistics models developped in this document on electric consumption data sets which were useful for the SOLENN project. A first application is linked to the electric constraint program directed by ENEDIS in order to secure the electric network. The houses are under a reduction of their maximal power for a short period of time. The goal is to study how the household behaves during this period of time. A second application concern the utilisation of the multiple regression model to study the effect of individuals visits on the electric consumption. The goal is to study the impact on the electric consumption for the week or the month following a visit
Corbière, Fabien. "Modèles de mélange en analyse de survie en présence de données groupées : application à la tremblante du mouton". Phd thesis, Université Victor Segalen - Bordeaux II, 2007. http://tel.archives-ouvertes.fr/tel-00195495.
Testo completoNous utilisons des modèles d'analyse des données de survie prenant en compte l'existence d'une fraction non à risque. Nous proposons une approche par vraisemblance pénalisée, qui allie les avantages des modèles paramétriques et semi paramétriques existants. Nous nous intéressons ensuite aux modèles paramétriques de survie avec fraction non à risque et effets aléatoires afin de tenir compte du regroupement des animaux dans les élevages. Ces différentes approches sont évaluées à l'aide d'études de simulations.
L'application des ces modèles aux données issues du suivi longitudinal d'élevages infectés des Pyrénées Atlantiques (France) confirme le rôle déterminant du génotype au gène PRP sur le risque de contamination et les durées d'incubation. Nos résultats suggèrent de plus que la contamination par l'agent infectieux a principalement lieu en période néonatale. Enfin la forte hétérogénéité des risques de contamination et des durées d'incubation mise en évidence entre troupeaux pourrait être partiellement expliquée par la prise en compte de la structure génétique des élevages et du nombre d'animaux infectés présents.
Colas, Sandrine. "Etude des déterminants de la survie prothétique des prothèses de hanche en France, à partir des données du SNIIRAM". Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLS421/document.
Testo completoMore and more Total Hip Replacement (THR) are performed in France (PTH) these last decades ; a 10% increase in 4 years has been observed, with 100 000 implantations on 2013.This increase can is related to the ageing of the population as well as the expansion of the implanted population: on one side younger and more active patients are now implanted, and on the other side, older patients (more than 80), often with other comorbidities, are now implanted.Le number of implanted THR, their characteristics, the context of implantation et the characteristics of the implanted population in France have never been comprehensively described so far. Some data are available from international registries (from Nordic, British, North American coutries and Australia), but not in France. The aim of my PhD was to study which factors were associated with the THR survivorship. My research covered the characteristics of the implants, of the patients and of the implanting centers.The data we used were from the French National Health Insurance Information System, SNIIRAM (Système National d'Information Inter-Régimes de l'Assurance Maladie) from 2006 to 2016. About 100 000 THR are implanted annually in France; the total cohort of THR implanted patients between 2006 and 2014 comprises about a million subjects. The cohorts studied in my work comprises between 100 000 and 300 000 patients, included between 2009 and 2012 and followed until 2013 to 2014.These cohorts studies showed that the implants characteristics, bearing surface, cementation, choice of a modular neck are associated with the prosthetic survivorship at short and midterm follow-up. We found the same with the patients characteristics, age, sex, diabetes mellitus as well as exposure to drugs such as benzodiazepines. The characteristics if the implanting center are also associated to the prosthetic survivorship, more specifically the volume of activity of the surgeon who performed the hip arthroplasty.The PMSI data are a valuable tool to perform an active surveillance of THR survivorship and using them allowed us to identify several risk factors of revision at short to midterm follow-up. About 100 000 patients receive a THR annually in France and THR revision is a surgical operation far more complicated than primary implantation, with higher complications rate during and post-operation. Being able to identifiy the factors associated with THR revision and being able to take the measures improving the THR survivorship are indeed a major Public health challenges. Our work' purpose is to assess the current practices and to provide evidences to promote technical choices propitious to THR survivorship, to contribute to health quality improvement
Touraine, Celia. "Modèles illness-death pour données censurées par intervalle : application à l'étude de la démence". Thesis, Bordeaux 2, 2013. http://www.theses.fr/2013BOR22099/document.
Testo completoIn dementia research, difficulties arise when studying cohort data. Time-to-disease onset is interval censored because the diagnosis is made at intermittent follow-up visits. As a result, disease status at death is unknown for subjects who are disease-free at the last visit before death. The illness-death model allows initially disease-free subjects to first become ill and then die, or die directly. Those two possible trajectories of the subjects who died without dementia diagnosis can be taken into account into the likelihood. Unlike the case where transition times are exactly observed, the latter do not factorizes and parameters of the three transitions have to be estimated jointly. However, when studying risk factors of dementia, a common approach consists in artificially ending follow-up of subjects who died without dementia diagnosis by considering them as right censored at the last time they were seen without disease. The first part of the present work shows that this approach (unlike the illness-death modeling approach) can lead to biases when estimating risk factor effects of dementia. Modeling death in addition to disease also allows to consider quantities which are closely related with risk of death, like lifetime risk of disease or life expectancies. In the second part of this work, we detail all the quantities which are of epidemiological interest in an illness-death model. They can be estimated, in addition to the transition intensities and the effects or risk factors, using the R package SmoothHazard which has been implemented during this thesis. Finally, in the last part of this work, we consider shared frailty regression models for the three transitions of the illness-death model
Parent, Contal Cécile. "AAdéquation et extension du modèle à risques proportionnels dans l'analyse des données de survie. Application au cancer du sein". Paris 6, 2000. http://www.theses.fr/2000PA066359.
Testo completoFromholtz, Raphaël. "Etude des supernovae de type Ia dans leur environnement à l'aide du SuperNova Legacy Survey et des données du COSMic evOlution Survey". Thesis, Aix-Marseille 2, 2010. http://www.theses.fr/2010AIX22087/document.
Testo completoOver the past decade supernovae have emerged as one of the most powerful tools for reconstructing the global history of the Universe. However type Ia supernovae are still empirical tools. Future experiments, as JDEM, are planned to better characterize the equation of state of the dark energy leading to the observed acceleration thousands of objects. These experiments will need to carefully control systematic errors to ensure future conclusions are not dominated by effects unrelated to cosmology. The evolution of N Ia with redshift or the presence of subclasses among them can be at the origin of that kind os systematics. So a better understanding of the properties of supernovae in their host galaxies could provide information about the correlation between supernovae and their environment, a better understanding of their standardization, finally a better description of supernovae as astrophisical object. this study can also provide informations for more realistic simultations of a space mission like JDEM
Beaudoin, David. "Estimation de la dépendance et choix de modèles pour des données bivariées sujettes à censure et à troncation". Thesis, Université Laval, 2007. http://www.theses.ulaval.ca/2007/24621/24621.pdf.
Testo completoLatouche, Aurélien. "Modèles de régression en présence de compétition". Paris 6, 2004. https://tel.archives-ouvertes.fr/tel-00129238.
Testo completoTessier, Maxime. "Étude de la performance d’un test d’association génétique pour des données familiales de survie en présence d’un biais de sélection". Master's thesis, Université Laval, 2020. http://hdl.handle.net/20.500.11794/66872.
Testo completoIn Leclerc et al. (2015, Genetic Epidemiology, 39 (6), 406-414), an association test between a group of genetic variants and censored phenotypes in presence of intrafamilial correlation is proposed. This test was implemented in a R package named gyriq. In this master’s thesis,we evaluate, with simulations, the performance of this test in presence of a sampling bias which stems from the data collection protocol. Indeed, in many situations, medical data from a family are considered if and only if a particular member of this family, called proband, is diagnosed with the event of interest during his medical exam. We develop multiple strategies to generate biased data according to such data collection protocol. We examine type 1 error and power of the association test in presence of such data, in the cases where there are 1 or more probands and when we sample only families where the probands have the event of interest or when we also sample a small proportion of families where the event has not occured for the probands. We conclude that the association test remains valid in presence of a selection bias but that the test power is diminished. Furthermore, the test is not valid when we include families where the event of interest has not occured for the probands.
Péron, Julien. "Evaluer le bénéfice clinique dans les essais randomisés en utilisant les comparaisons par paire généralisées incluant des données de survie". Thesis, Lyon 1, 2015. http://www.theses.fr/2015LYO10190/document.
Testo completoIn medical oncology randomized trials, treatment effect is usually assessed on several endpoints, including one or more time-to-event endpoints. An overall analysis of the treatment effect may include the outcomes observed on all the relevant endpoints. A systematic review of medical oncology phase III trials was conducted. We extracted the methods used to record, analyze and report adverse events and patient-reported outcomes. Our findings show that some methodological aspects of adverse events or patient-reported outcomes collection and analysis were poorly reported. Even when reported, the methods used were highly heterogeneous. Another objective was to develop an extension of the generalized pairwise comparison procedure for time-to-event variables. The extended procedure provides an unbiased estimation of the chance of a better outcome even in presence of highly censored observations. Then, we show how the chance of an overall better outcome can be used to assess the benefit-risk balance of treatment in randomized trials. When a benefit is expected on more than one endpoint, the chance of an overall better outcome assesses the overall therapeutic benefit. The test of the null hypothesis is more powerful than the test based on one single endpoint
Ternes, Nils. "Identification de biomarqueurs prédictifs de la survie et de l'effet du traitement dans un contexte de données de grande dimension". Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLS278/document.
Testo completoWith the recent revolution in genomics and in stratified medicine, the development of molecular signatures is becoming more and more important for predicting the prognosis (prognostic biomarkers) and the treatment effect (predictive biomarkers) of each patient. However, the large quantity of information has rendered false positives more and more frequent in biomedical research. The high-dimensional space (i.e. number of biomarkers ≫ sample size) leads to several statistical challenges such as the identifiability of the models, the instability of the selected coefficients or the multiple testing issue.The aim of this thesis was to propose and evaluate statistical methods for the identification of these biomarkers and the individual predicted survival probability for new patients, in the context of the Cox regression model. For variable selection in a high-dimensional setting, the lasso penalty is commonly used. In the prognostic setting, an empirical extension of the lasso penalty has been proposed to be more stringent on the estimation of the tuning parameter λ in order to select less false positives. In the predictive setting, focus has been given to the biomarker-by-treatment interactions in the setting of a randomized clinical trial. Twelve approaches have been proposed for selecting these interactions such as lasso (standard, adaptive, grouped or ridge+lasso), boosting, dimension reduction of the main effects and a model incorporating arm-specific biomarker effects. Finally, several strategies were studied to obtain an individual survival prediction with a corresponding confidence interval for a future patient from a penalized regression model, while limiting the potential overfit.The performance of the approaches was evaluated through simulation studies combining null and alternative scenarios. The methods were also illustrated in several data sets containing gene expression data in breast cancer
Bouatou, Mohamed. "Estimation non linéaire par ondelettes : régression et survie". Phd thesis, Université Joseph Fourier (Grenoble), 1997. http://tel.archives-ouvertes.fr/tel-00004921.
Testo completoBruandet, Amélie. "Facteurs pronostiques de patients atteints de démence suivis en Centre mémoire de ressources et de recherche : exemple d'utilisation de bases de données médicales à des fins de recherche clinique". Lille 2, 2008. http://tel.archives-ouvertes.fr/tel-00336252/fr/.
Testo completoFereres, Yohan. "Stratégies d’arbitrage systématique multi-classes d'actifs et utilisation de données hétérogènes". Thesis, Paris Est, 2013. http://www.theses.fr/2013PEST0075/document.
Testo completoFinancial markets evolve more or less rapidly and strongly to all kind of information depending on time period of study. In this context, we intend to measure a broad set of information influence on systematic multi-assets classes “euro neutral” arbitrage portfolios either for “naive” diversification and optimal diversification. Our research focuses on systematic tactical asset allocation and we group these information under the name of heterogeneous data (market data and “other market information”). Market data are “end of day” asset closing prices and “other market information” gather economic cycle, sentiment and volatility indicators. We assess the influence of a heterogeneous data combination on our arbitrage portfolios for a time period including the subprimes crisis period and thanks to data analysis and quantization algorithms. The impact of a heterogeneous data combination on our arbitrage portfolio is materialized by increasing return, increasing return/volatility ratio for the post subprimes crisis period, decreasing volatility and asset class correlations. These empirical findings suggest that “other market information” presence could be an element of arbitrage portfolio risk diversification. Furthermore, we investigate and bring empirical results to Blitz and Vliet (2008) issue on global tactical asset allocation (GTAA) by considering “predictive” variables with a systematic market timing process integrating heterogeneous data thanks to a quantitative data processing
Corbière, Fabien. "Modèles de mélange en analyse de survie en présence de donnée groupées : application à la tremblante du mouton". Bordeaux 2, 2007. http://www.theses.fr/2007BOR21465.
Testo completoAlthough they are of major interest in the understanding of the classical scrapie dynamic in infested flocks, the factors influencing the contamination by the infectious agent and incubation period remain poorly known. The absence of antemortem diagnostic tools and the confounding effects of long incubation periods and flock management practices yield a partial knowledge of the infectious status of animals. Moreover, we must take into account that only an unknown fraction of animals become infected, even in heavily affected flocks. To deal with issues, we use a class of particular survival models, which take account for the presence of long term survivors. We propose a penalized likelihood, allowing for the estimation of a smooth risk function. We also develop some parametric models with shared frailties to deal with the presence of grouped data. These different models are evaluated through simulations studies. These statistical approaches are then applied to the analysis of real data collected during the following-up of infected flocks in Pyrénées Atlantiques. The key influence of the PRP genotype on the contamination risk and incubation periods is confirmed. Our results also suggest that, at the individual level, the infection mainly takes place around birth. Finally, the strong heterogeneity in the contamination risk and incubation periods observed between flocks could be partially explained by their PRP genetic structure and the number of incubating animals
Latouche, Aurélien. "Modèles de regression en présence de compétition". Phd thesis, Université Pierre et Marie Curie - Paris VI, 2004. http://tel.archives-ouvertes.fr/tel-00129238.
Testo completoCe travail de thèse a porté sur l'étude de modèles de régression dans ce cadre. Deux approches ont été envisagéés, l'une basée sur la fonction de risque instantané cause-spécifique et l'autre sur la
fonction de risque instantané associée à la fonction d'incidence cumulée (ou fonction de risque de sous-répartition). Dans les deux cas, nous avons considéré une formulation à risques proportionnels, c'est à dire un modèle de Cox dans le premier cas, et un modèle de Fine & Gray dans le second cas.
Dans un premier temps, l'implication d'un choix de modélisation a été étudiée dans le cadre de la planification d'un essai clinique ou d'une étude pronostique. Nous avons alors developpé
une formule de calcul du nombre de sujets nécessaire pour le modèle de Fine & Gray.
Puis, les conséquences d'un modèle mal spécifié pour la fonction de risque de sous répartition ont été evaluées, en étudiant l'influence sur l'estimation du paramètre de régression dans le modèle de Fine & Gray quand le vrai modèle est un modèle à risques proportionnels pour la fonction de risque cause-
spécifique. Nous avons ensuite étudié les implications de l'inclusion de covariables dépendantes du
temps dans le modèle de Fine & Gray.
Enfin, nous avons présenté une synthèse didactique sur la
stratégie d'utilisation des approches présentées dans ce travail de thèse.
Tchatchueng, Mbougua Jules Brice. "Problématiques statistiques rencontrées dans l’étude du traitement antirétroviral des adultes infectés par le VIH en Afrique subsaharienne". Thesis, Montpellier 1, 2012. http://www.theses.fr/2012MON1T006/document.
Testo completoOn the basis of statistical challenges encountered in study of antiretroviral treatment of adults infected with human immunodeficiency virus (HIV) in sub-Saharan Africa, this thesis aims to promote the dissemination of relatively recent methodological tools of less aware audience of users on one hand and to participate to development of new tools on the other hand. The first chapter presents various methods for modeling longitudinal data of which analysis methods for changing of a criterion over time (the generalized linear mixed models and models of generalized estimating equations) or the occurrence of an event over time (the semi-parametric Cox model and its extensions to take into account time-dependent covariates and informative censoring). The second chapter focuses on non-inferiority test and provides two developments of the classical procedure of these tests in cases where the non-inferiority margin is relative. The third chapter addresses the question of missing data and proposes an extension of the multiple imputation method based on fully conditional specification, to take into account nonlinear effects of covariates in the imputation models using B-splines functions. These methods are illustrated by studies on HIV in Cameroon and Senegal
Frévent, Camille. "Contribution to spatial statistics for high-dimensional and survival data". Electronic Thesis or Diss., Université de Lille (2022-....), 2022. http://www.theses.fr/2022ULILS032.
Testo completoIn this thesis, we are interested in statistical spatial learning for high-dimensional and survival data. The objective is to develop unsupervised cluster detection methods by means of spatial scan statistics in the contexts of functional data analysis in one hand and survival data analysis in the other hand. In the first two chapters, we consider univariate and multivariate functional data measured spatially in a geographical area. We propose both parametric and nonparametric spatial scan statistics in this framework. These univariate and multivariate functional approaches avoid the loss of information respectively of a univariate method or a multivariate method applied on the average of the observations during the study period. We study the new methods' performances in simulation studies before applying them on economic and environmental real data. We are also interested in spatial cluster detection of survival data. Although there exist already spatial scan statistics approaches in this framework in the literature, these do not take into account a potential correlation of survival times between individuals of the same spatial unit. Moreover, the spatial nature of the data implies a potential dependence between the spatial units, which should be taken into account. The originality of our proposed method is to introduce a spatial scan statistic based on a Cox model with a spatial frailty, allowing to take into account both the potential correlation between the survival times of the individuals of the same spatial unit and the potential dependence between the spatial units. We compare the performances of this new approach with the existing methods and apply them on real data corresponding to survival times of elderly people with end-stage kidney failure in northern France. Finally, we propose a number of perspectives to our work, both in a direct extension of this thesis in the framework of spatial scan statistics for high-dimensional and survival data, but also perspectives in a broader context of unsupervised spatial analysis (spatial clustering for high-dimensional data (tensors)), and supervised spatial learning (regression)
Soret, Perrine. "Régression pénalisée de type Lasso pour l’analyse de données biologiques de grande dimension : application à la charge virale du VIH censurée par une limite de quantification et aux données compositionnelles du microbiote". Thesis, Bordeaux, 2019. http://www.theses.fr/2019BORD0254.
Testo completoIn clinical studies and thanks to technological progress, the amount of information collected in the same patient continues to grow leading to situations where the number of explanatory variables is greater than the number of individuals. The Lasso method proved to be appropriate to circumvent over-adjustment problems in high-dimensional settings.This thesis is devoted to the application and development of Lasso-penalized regression for clinical data presenting particular structures.First, in patients with the human immunodeficiency virus, mutations in the virus's genetic structure may be related to the development of drug resistance. The prediction of the viral load from (potentially large) mutations allows guiding treatment choice.Below a threshold, the viral load is undetectable, data are left-censored. We propose two new Lasso approaches based on the Buckley-James algorithm, which imputes censored values by a conditional expectation. By reversing the response, we obtain a right-censored problem, for which non-parametric estimates of the conditional expectation have been proposed in survival analysis. Finally, we propose a parametric estimation based on a Gaussian hypothesis.Secondly, we are interested in the role of the microbiota in the deterioration of respiratory health. The microbiota data are presented as relative abundances (proportion of each species per individual, called compositional data) and they have a phylogenetic structure.We have established a state of the art methods of statistical analysis of microbiota data. Due to the novelty, few recommendations exist on the applicability and effectiveness of the proposed methods. A simulation study allowed us to compare the selection capacity of penalization methods proposed specifically for this type of data.Then we apply this research to the analysis of the association between bacteria / fungi and the decline of pulmonary function in patients with cystic fibrosis from the MucoFong project
Jardillier, Rémy. "Evaluation de différentes variantes du modèle de Cox pour le pronostic de patients atteints de cancer à partir de données publiques de séquençage et cliniques". Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALS008.
Testo completoCancer has been the leading cause of premature mortality (death before the age of 65) in France since 2004. For the same organ, each cancer is unique, and personalized prognosis is therefore an important aspect of patient management and follow-up. The decrease in sequencing costs over the last decade have made it possible to measure the molecular profiles of many tumors on a large scale. Thus, the TCGA database provides RNA-seq data of tumors, clinical data (age, sex, grade, stage, etc.), and follow-up times of associated patients over several years (including patient survival, possible recurrence, etc.). New discoveries are thus made possible in terms of biomarkers built from transcriptomic data, with individualized prognoses. These advances require the development of large-scale data analysis methods adapted to take into account both survival data (right-censored), clinical characteristics, and molecular profiles of patients. In this context, the main goal of the thesis is to compare and adapt methodologies to construct prognostic risk scores for survival or recurrence of patients with cancer from sequencing and clinical data.The Cox model (semi-parametric) is widely used to model these survival data, and allows linking them to explanatory variables. The RNA-seq data from TCGA contain more than 20,000 genes for only a few hundred patients. The number p of variables then exceeds the number n of patients, and parameters estimation is subject to the “curse of dimensionality”. The two main strategies to overcome this issue are penalty methods and gene pre-filtering. Thus, the first objective of this thesis is to compare the classical penalization methods of Cox's model (i.e. ridge, lasso, elastic net, adaptive elastic net). To this end, we use real and simulated data to control the amount of information contained in the transcriptomic data. Then, the second issue addressed concerns the univariate pre-filtering of genes before using a multivariate Cox model. We propose a methodology to increase the stability of the genes selected, and to choose the filtering thresholds by optimizing the predictions. Finally, although the cost of sequencing (RNA-seq) has decreased drastically over the last decade, it remains too high for routine use in practice. In a final section, we show that the sequencing depth of miRNAs can be reduced without degrading the quality of predictions for some TCGA cancers, but not for others
Marti, soler Helena. "Modélisation des données d'enquêtes cas-cohorte par imputation multiple : Application en épidémiologie cardio-vasculaire". Phd thesis, Université Paris Sud - Paris XI, 2012. http://tel.archives-ouvertes.fr/tel-00779739.
Testo completoDoux, Cyrille. "Combinaisons de sondes cosmologiques : deux applications avec les données de Planck et SDSS-III/BOSS". Thesis, Sorbonne Paris Cité, 2017. http://www.theses.fr/2017USPCC230/document.
Testo completoThis thesis addresses the combinations of cosmological probes from the measurements of the cosmic microwave background (CMB) and galaxy redshift surveys, and exploits data from the Planck satellite and the Baryon Oscillation Spectroscopic Survey (BOSS) of the Sloan Digital Sky Survey. It explores how cross-correlations between different data sets can be used to detect new signals and improve contraints on cosmological parameters. First, we measure, for the first time, the cross-correlation between gravitational lensing of the CMB and the power spectrum of the Lyman-α forest in the spectra of quasars. This effect, which emerges from purely non-linear evolution, is interpreted as the response of the power spectrum to large-scale modes. It shows how fluctuations in the density of neutral hydrogen in the intergalactic medium are affected by large-scale fluctuations in the density of dark matter. The measured signal is compatible with the theoretical approach and simulations run by another group. In a second time, we develop a formalism enabling the joint analysis of the galaxy/quasar density contrast and CMB lensing. Taking cross-correlations between these probes into account reduces error bars on some cosmological parameters by up to 20%, equivalent to an increase in the size of the survey of about 50%. This analysis is completed by CMB temperature anisotropies information in order to constrain all the parameters of the ΛCDM standard model and galaxy biases at once. Finally, it is extended to obtain contraints on the dark energy equation of state as well as the sum of the masses of neutrinos
Younes, Hassan. "Estimation du taux de mortalité sous contraintes d'ordre pour des données censurées ou tronquées /". Montréal : Université du Québec à Montréal, 2005. http://accesbib.uqam.ca/cgi-bin/bduqam/transit.pl?&noMan=24065971.
Testo completoBruandet, Amelie. "Facteurs pronostiques de patients atteints de démence suivis en centre mémoire de ressource et de recherche : exemple d'utilisation de bases de données médicales à des fins de recherche clinique". Phd thesis, Université du Droit et de la Santé - Lille II, 2008. http://tel.archives-ouvertes.fr/tel-00336252.
Testo completoL'objectif de mon travail est l'étude des facteurs pronostiques de patients, pris en charge au centre de mémoire de ressource et de recherche (CMRR) du Centre Hospitalier Régional et Universitaire (CHRU) de Lille et du centre médical des monts de Flandres de Bailleul. Pour cela, nous avons utilisé la base de données médicales informatisées des patients consultant au CMRR de Lille-Bailleul. Ce travail s'est en particulier intéressé aux avantages et aux limites de l'utilisation de bases de données médicales à des fins de recherche clinique dans l'étude des facteurs pronostiques des démences.
Dans une population de 670 patients ayant une maladie d'Alzheimer, nous avons confirmé que le déclin des fonctions cognitives (évaluées par le MMSE) était significativement plus élevé chez les sujets ayant un niveau d'éducation intermédiaire ou élevé par rapport aux sujets ayant un bas niveau d'éducation. Cependant, la mortalité ne différaient pas de façon significative entre ces trois groupes. Nous avons décrit une mortalité similaire entre patients ayant une maladie d'Alzheimer, une démence mixte ou une démence vasculaire. Les patients ayant une démence mixte avaient un déclin du MMSE plus important que les patients ayant une démence vasculaire mais moins important que les patients ayant une maladie d'Alzheimer. Enfin, nous avons montré que le risque de développer une démence vasculaire ou mixte augmentait de manière significative avec le nombre d'hypersignaux sous corticaux chez des patients ayant un mild cognitive impairment.
Ces travaux soulignent la difficulté de l'établissement du diagnostic des démences mixtes, la complexité de l'analyse du déclin des fonctions cognitives (prise en compte du stade de progression des démences, absence d'instrument de suivi des fonctions cognitives à la fois simple d'utilisation et sensible aux faibles variations au cours du temps ou non linéarité du déclin des fonctions cognitives), les avantages en terme de coût et de temps de l'utilisation de bases de données médicales, et les problème de sélection de la population issue d'une structure de soins.
Malgré les problèmes de représentativité des populations, ce travail montre l'intérêt de l'utilisation à des fins de recherche clinique de données médicales concernant des patients pris en charge en structure de soins.