Doctorant F/H Impact of treatment administration on the progression of neurodegenerative diseases

Télécharger l'offre au format PDF

Type de contrat : CDD

Niveau de diplôme exigé : Bac + 5 ou équivalent

Fonction : Doctorant

Contexte et atouts du poste

Treatment effects on disease progression are key elements to support therapeutic decisions. Methods exist to model the natural progression of the disease. Among them, the Disease Course Mapping method proposed by Schiratti et al. [1] allows modeling individual trajectories on Riemannian manifolds within a Bayesian mixed–effects framework and is implemented in an open-source software library called Leaspy. 

Modeling how a treatment influences this evolution is challenging. One option is to use an extension of the Disease Course Mapping toward a piecewise-geodesic formulation. This extension allows for capturing structural breaks in disease progression, such as those potentially induced by therapeutic interventions. A compelling methodological basis for this type of model has been proposed by Chevallier et al. [2], who applied it to study treatment effects in kidney cancer. While the model shows promise, fundamental questions about its identifiability and practical implementation remain to be addressed. This is particularly true for settings involving complex parameterizations or sparse data. 

The objective of this PhD project is to investigate these issues, implement them in the Leaspy library if feasible or in a new library, and apply the method to real data from patients affected by neurodegenerative disease. 

 References: 

[1] Schiratti, Jean-Baptiste, Stéphanie Allassonnière, Olivier Colliot, and Stanley Durrleman. "A Bayesian mixed-effects model to learn trajectories of changes from repeated manifold-valued observations." Journal of Machine Learning Research18, no. 133 (2017): 1-33. 

[2] Juliette Chevallier, Stéphane Oudard, Stéphanie Allassonnière. Learning spatiotemporal piecewise geodesic trajectories from longitudinal manifold-valued data. 31st Conference on Neural Information Processing Systems (NIPS 2017), Dec 2017, Long Beach, United States. hal-01646230 

Mission confiée

  • investigate how to implement the effect of treatment on a longitudinal multivariate model,
  • implement them in the Leaspy library if feasible or in a new library, 
  • apply the method to real data from patients affected by neurodegenerative disease. 

Principales activités

Activités principales :

  • Veille scientifique et construction d'une bibliographie
  • Rédaction et publication d'articles scientiques
  • Implémentations d'artefacts logiciels effectuant la démonstration des résultats de recherche
  • Apprentissage des compétences du métier de la recherche

Activités secondaires :

  • Participation à la vie scientifique de l'équipe (séminaires, groupes de lecture)
  • Participation à des conférences scientifiques

Avantages

  • Restauration subventionnée
  • Transports publics remboursés partiellement
  • Congés: 7 semaines de congés annuels + 10 jours de RTT (base temps plein) + possibilité d'autorisations d'absence exceptionnelle (ex : enfants malades, déménagement)
  • Possibilité de télétravail et aménagement du temps de travail
  • Équipements professionnels à disposition (visioconférence, prêts de matériels informatiques, etc.)
  • Prestations sociales, culturelles et sportives (Association de gestion des œuvres sociales d'Inria)
  • Accès à la formation professionnelle
  • Sécurité sociale