Multi-Modal Representational Learning for Computational Histopathology
Strasbourg, Bas-Rhin
CDD
Temps-plein
Il y a 21 jours
Offer DescriptionThe advent of computational histopathology, facilitated by Digital Pathology and Whole-Slide Imaging (WSI), has revolutionized disease diagnosis and understanding. This transformation has led to the emergence of Computational Digital Pathology, promising faster and more accurate diagnoses, prognoses, and predictions of serious diseases. However, challenges arise from the vast amounts of image data, staining variability, and image heterogeneity associated with WSI. To address these challenges, this project aims to develop automated diagnostic approaches for histopathology images. The main objectives include:Developing novel methods to learn effective multi-modal representations from histopathological images and corresponding reports.Developing methods for segmenting multiple anatomical structures and cells using deep learning approaches.Overcoming scientific barriers by proposing new multi-class segmentation approaches, developing strategies for domain invariance, ensuring interpretability, and defining best practices for integrating information from multi-modal sources.RequirementsResearch Field Computer science Education Level Master Degree or equivalentSkills/QualificationsMaster's in computer science or equivalentPython/C++ programming skillsStrong knowledge of computer vision and machine learningProficiency in English (oral and written)Experience with Deep LearningLanguages ENGLISH Level ExcellentInternal Application form(s) neededResearch_PhD_HistoGraph.pdfEnglish(253.35 KB - PDF)Additional InformationWork Location(s)Number of offers available 1 Company/Institute University of Strasbourg, IHU Strasbourg Country France State/Province Bas-Rhin City Strasbourg Postal Code 67000 GeofieldWhere to apply E-mailnpadoy@unistra.frContact CityStrasbourg WebsiteStreet1, place de l'Hôpital Postal Code67091 Strasbourg Cedex E-Mailnpadoy@unistra.frsrivastav@unistra.frSTATUS: EXPIRED