Looking for postdocs/engineers in the field of deep learning for microscopy
CNRS@Create & the Mechanobiology Institute at NUS recruit 3 engineers/post-docs in the field of deep learning for a joint project of large-scale image processing and analysis.
The candidates will join an international team of about 10 people working on developing a new class of high content screening imaging platform for organoids. The project involves microfabrication, image acquisition, image analysis and high-content-screening. The candidates will be in charge of developing deep learning-based processing and analysis approaches to i) optimize the acquisition workflow, ii) segment organoids morphologies, and iii) predict organoid fates.
JOB DESCRIPTION AND MAIN TASKS
The candidates will be in charge of implementing and adapting AI algorithms to one of the three problems below:
- Image enhancement and denoising to limit the cytotoxicity of the detection and maximize the speed of imaging.
- Automated detection and classification of morphological features in the developing organoids in 3D (live and fixed).
- Prediction of organoid development fate based on the live evolution of the organoid morphologies.
This project results from a 7 year collaboration between the IINS in Bordeaux, the Mechanobiology Institute in Singapore and the CNRS BMC2 lab in Singapore. The proof of concept of high content screening capability of the platform have been achieved, with a current capability of 100 3D organoids per/hour. The consortium also achieved proof of concept that deep learning networks can be trained on the acquired images. In this context the candidates will directly interact with the experimental team. The project aims at developing a new class of imager for organoids that can be used not only in research labs but also as a quality control device and organoids testing device by the industry. Private/public partnership around this project are in final phase of being established, providing the candidate a chance to interact with the private sector. The project is jointly developed in Bordeaux and Singapore. The candidates will be offered a unique opportunity to work (according to their will) in both countries.
The candidates must:
- Have a PhD or equivalent in computer science or a related field,
- With proven experience in developing bioimage analysis tools.
- Knowledge and ability in one or more deep learning frameworks (Tensorflow, Keras, Torch, Caffe, etc), preferably in Python, is required.
- Previous experience working with medical or biological images is prefered.
24 to 36 months.