BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:8639@i2m.univ-amu.fr
DTSTART;TZID=Europe/Paris:20250512T093000
DTEND;TZID=Europe/Paris:20250512T123000
DTSTAMP:20250410T090701Z
URL:https://www.i2m.univ-amu.fr/evenements/exploration-de-lorganisation-ti
 ssulaire-et-des-dynamiques-cellulaires-stochastiques-dans-limagerie-a-faib
 le-flux-de-photons-dorganoides-diffusants-la-lumiere-strategies-c/
SUMMARY:Jules Vanaret (I2M): Exploration de l’organisation tissulaire et 
 des dynamiques cellulaires stochastiques dans l’imagerie à faible flux 
 de photons d’organoïdes diffusants la lumière  : stratégies combinant
  apprentissage profond et approches bayésiennes
DESCRIPTION:Jules Vanaret: Le jury sera composé de :\n- M. David ROUSSEAU\
 , Univ. d’Angers\, IRHS\, Rapporteur\n- M. Nicolas LOMENIE\, Univ. Paris
  Cité\, LIPADE\, Rapporteur\n- M. Hervé ISAMBERT\, CNRS\, Institut Curie
 \, Examinateur\n- M. Daniel SAGE\, EPFL\, Laboratoire d’imagerie bioméd
 icale\, Examinateur\n- M. Frédéric RICHARD\, Aix-Marseille Univ.\, I2M\,
  Directeur de thèse\n- M. Philippe ROUDOT\, Aix-Marseille Univ.\, Institu
 t Fresnel\, Co-directeur de thèse\n- Mme. Sham TLILI\, Aix-Marseille Univ
 .\, IBDM\, Co-encadrante de thèse\n\nAbstract:\n\nThis thesis explores th
 e biophysical and computational modeling of self-organization in gastruloi
 ds\, three-dimensional in vitro models of embryonic development. It invest
 igates how cellular decisions emerge from mechanogenetic interactions at t
 he single-cell level\, leveraging advanced microscopy\, image analysis\, a
 nd tracking methodologies to quantify cellular behaviors and micro-environ
 mental cues.\n\nThe first part presents a methodological pipeline to quant
 ify mechanical and genetic properties in deep gastruloid tissues. A deep-l
 earning-based segmentation strategy tailored for heterogeneous nuclei shap
 es is then developed\, integrating contrast enhancement and dual-view fusi
 on to improve imaging depth and accuracy. Building on these segmentation 
  advances\, the study probes the mechanogenetic landscape of gastruloids\,
  integrating quantitative analyses of gene expression and cellular deforma
 tion. A dedicated Python-based  computational pipeline is introduced\, en
 abling multiscale analysis and user-friendly visualization.\n\nIn the seco
 nd part\, the challenges of cell tracking in gastruloids are assessed\, pr
 esenting how stochasticity and measurement uncertainty complicate lineage 
 reconstruction. Various tracking methodologies applied in the context of c
 ell tracking or developmental biology are reviewed\, leading to the develo
 pment of a Bayesian filtering framework for the multiscale quantification 
 of tracking uncertainty.\n\nThe final sections apply these probabilistic m
 ethods to validate segmentation and tracking algorithms\, demonstrating ho
 w the Bayesian filtering algorithm can guide unsupervised parameter tuning
  and segmentation algorithm ranking. Through extensive simulations and exp
 erimental applications\, the study provides rigorous methods to aid cellul
 ar dynamics analysis in complex gastruloid datasets\, with implications fo
 r developmental biology and tissue engineering.\n\nKeywords: embryonic org
 anoid\, cell tracking\, live imaging\, cellular heterogeneity\, deep-learn
 ing\, bayesian filtering
CATEGORIES:Soutenance de thèse,Signal et Apprentissage
LOCATION:Luminy - Auditorium de l'Hexagone\, 172 avenue de Luminy\, Marseil
 le\, 13009\, France
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=172 avenue de Luminy\, Mars
 eille\, 13009\, France;X-APPLE-RADIUS=100;X-TITLE=Luminy - Auditorium de l
 'Hexagone:geo:0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:DAYLIGHT
DTSTART:20250330T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR