PostDoc for Segmentation in Virtual Environment

PostDoc for Segmentation in Virtual Environment

The DIVA: Data Integration and Visualization in Virtual and Augmented environment aims at creating an intuitive environment where data treatment leverages Virtual Reality (VR), human data interaction and automated algorithmic treatment. It is a joined initiative between the Pasteur institute and the Curie Institute. Key to this approach is the relation between operator intervention and data treatment. This approach is tailored to handle data in limited quantities, possibly very noisy, and to allow quick robust learning of the required procedures to extract relevant information from data.

A software platform DIVA has already been designed allowing any image stack, either from microscopy or from MRI/CT scans to be automatically integrated in a VR environment. The platform generates an entire environment allowing interaction with the data as well as changes in the representation in 3D and active modification/annotation/projection of the data. The platform is at the moment only mainly available in alpha version in both Pasteur and Curie Institutes.

More information on applications of the DIVA project is available at the following address: https://research.pasteur.fr/en/project/data-integration-and-visualisation-in-augmented-and-virtual-environments/


We are proposing a 1-year postdoctoral fellowship (renewable for 1 year) financed by the DIM-ELICIT for a motivated applicant. She/he is expected to contribute to our efforts of designing segmentation procedure for difficult noisy data coming from microscopy images and biomedical images. The envisioned approach will join Bayesian Pixel features learning (within the VR environment) and amortized active membrane segmentation to direct interaction with the data by the user in the VR environment. Strong experience in programming is required and previous experience in pixel classification and segmentation will be appreciated. The approach will be tested on large-scale multi modal recordings of neuronal trees (light sheet, 2 photons and pathology) and if times allow on CT scans of breast and hepatic cancer.

The successful candidate will be working directly with both designers and coders of the DIVA platform: Mohamed El Beheiry and Sebastien Doutreligne. She/He will be working directly with various members of the Lledo laboratory to test the approaches on experimental data. She/he will join a highly multidisciplinary lab where physicists and computer scientists combine approaches to question random walks of biomolecule in the cells and study relation between neural architecture and behavior in the drosophila larva. Finally, the successful candidate will have privileged access to multiple GPU clusters in the Pasteur institute the CPU cluster and custom platforms in the lab.


Required Skills

  • For a physicist: statistical physics or condensed matter physics.
  • For a computer scientist: Bayesian Inference, variational inferences, segmentation and optimization
  • Scripting language experience: Bash, Python

Nice to have Skills

  • pytorch

Duration

1-2 years

Localization

Institut Pasteur (Paris)

Contact Information

Interested candidates should contact Jean-Baptiste Masson (jbmasson@pasteur.fr) and Maxime Dahan (Maxime.Dahan@curie.fr) with their CV, a brief motivation letter and 2 letters of recommendation.

2 Likes