Cette thèse présente de nouveaux outils informatiques pour la modélisation conjointe de données biomédicales multimodales, robustes aux données manquantes, avec une application aux études de neuro-imagerie dans les maladies neurodégénératives. La base théorique de notre travail est l'auto-encodeur variationnel (VAE), un modèle de variables latentes bien adapté pour travailler avec des données complexes car il les projette dans un espace plus simple et de faible dimension, capable de modéliser les non-linéarités des données. Le cœur de cette thèse consiste en l'autoencodeur variationnel multicanal (MCVAE), une extension du VAE pour modéliser conjointement les relations latentes entre les observations multimodales. Ceci est réalisé 1) en cont...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis presents new computational tools for the joint modeling of multi-modal biomedical data,r...
The joint modeling of neuroimaging data across multiple datasets requires to consistently analyze hi...
International audienceInterpretable modeling of heterogeneous data channels is essential in medical ...
The joint modeling of neuroimaging data across multiple datasets requires to consistently analyze hi...
International audienceInterpretable modeling of heterogeneous data channels is essential in medical ...
International audienceInterpretable modeling of heterogeneous data channels is essential in medical ...
PosterInternational audienceThe aim of this study is to develop a generative and probabilistic stati...
PosterInternational audienceThe aim of this study is to develop a generative and probabilistic stati...
International audienceThe joint analysis of biomedical data in Alzheimer's Disease (AD) is important...
International audienceThe joint analysis of biomedical data in Alzheimer's Disease (AD) is important...
International audienceThe joint analysis of biomedical data in Alzheimer's Disease (AD) is important...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis presents new computational tools for the joint modeling of multi-modal biomedical data,r...
The joint modeling of neuroimaging data across multiple datasets requires to consistently analyze hi...
International audienceInterpretable modeling of heterogeneous data channels is essential in medical ...
The joint modeling of neuroimaging data across multiple datasets requires to consistently analyze hi...
International audienceInterpretable modeling of heterogeneous data channels is essential in medical ...
International audienceInterpretable modeling of heterogeneous data channels is essential in medical ...
PosterInternational audienceThe aim of this study is to develop a generative and probabilistic stati...
PosterInternational audienceThe aim of this study is to develop a generative and probabilistic stati...
International audienceThe joint analysis of biomedical data in Alzheimer's Disease (AD) is important...
International audienceThe joint analysis of biomedical data in Alzheimer's Disease (AD) is important...
International audienceThe joint analysis of biomedical data in Alzheimer's Disease (AD) is important...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...
This thesis focuses on the statistical learning of digital models of neurodegenerative disease progr...