Séminaires de Biostatistique 2020

Ces évènement du BPH, Bordeaux Population Health – UMR 1219 sont organisés sous l’égide du Département Santé publique de l’université de Bordeaux.


Jeudi 13 février 2020 – 13h à 14h

Bayesian sparsity for statistical learning in high dimensions: applications to medical imaging

Amphi Pierre-Alexandre Louis, ISPED
Campus Carreire, 146 rue Léo Saignat, 33076 Bordeaux
entrée libre, sans inscription

Intervenant : Charles Bouveyron  (https://math.unice.fr/~cbouveyr/)

Résumé :

Although the ongoing digital revolution in fields such as chemometrics, genomics or personalized medicine gives hope for considerable progress in these areas, it also provides more and more high-dimensional data to analyze and interpret. A common usual task in those fields is discriminant analysis, which however may suffer from the high dimensionality of the data. The recent advances, through subspace classifica- tion or variable selection methods, allowed to reach either excellent classification performances or useful visualizations and interpretations. Obviously, it is of great interest to have both excellent classification accuracies and a meaningful variable selection for interpretation. This work addresses this issue by intro- ducing a subspace discriminant analysis method which performs a class-specific variable selection through Bayesian sparsity. The resulting classification methodology is called sparse high-dimensional discrimi- nant analysis (sHDDA). Contrary to most sparse methods which are based on the Lasso, sHDDA relies on a Bayesian modeling of the sparsity pattern and avoids the painstaking and sensitive cross-validation of the sparsity level. The main features of sHDDA are illustrated on simulated and real-world data. In particular, we propose an exemplar application to cancer characterization based on medical imaging using radiomic feature extraction is in particular proposed.
[F. Orlhac, P.-A. Mattei, C. Bouveyron and N. Ayache, Class-specific Variable Selection in High-Dimensional Discriminant Analysis through Bayesian Sparsity, Journal of Chemometrics, vol. 33, pp. e3097, 2019] [C. Bouveyron, P. Latouche and P.-A. Mattei, Bayesian Variable Selection for Globally Sparse Probabilistic PCA, Electronic Journal of Statistics, vol. 12(2), pp. 3036-3070, 2018]

Programme des prochains séminaires de Biostatistique :

– mardi 18 février : Jérémie Lespinasse (17h à 18h Salle Chastang à l’ISPED – campus Carreire) sur la « modélisation de la séquence temporelle des atteintes dans la cohorte MEMENTO par modèle mixte multivarié avec recalage temporel« 
– jeudi 5 mars : Arnaud Gloaqen (13h à 14h Amphi Louis à l’ISPED – campus de Carreire)
– jeudi 23 avril : Lynn Lin (13h à 14h Amphi Louis à l’ISPED – campus de Carreire)
– jeudi 30 avril : Christopher Foley (13h à 14h Amphi Louis à l’ISPED – campus de Carreire)
– jeudi 4 juin : Paul Kirk (13h à 14h Amphi Louis à l’ISPED – campus de Carreire)
– jeudi 25 juin : Stacia M DeSantis (13h à 14h Amphi Louis à l’ISPED – campus de Carreire)

Retrouvez tous nos séminaires sur la page dédiée : https://www.bordeaux-population-health.center/nos-seminaires/

logo_ISPED logo_SFR-santepublique-societe logo_CHU-Bordeaux logo_Institut-Bergonie logo_INRIA logo_ANRS logo_ARS logo_Charles-Perrens logo_IFSTTAR logo_HONcode