PhD Research Fellow in Deep learning on image data (ref 257213)

Two phd positions are open. They are both focused on deep learning for images. Position 1 focuses on medical ultrasound images. Position 2 focuses on image data for monitoring marine ecosystems.  You must indicate in you application if you apply for both projects, or if you apply only for one of the projects. State project 1 for medical images, and project 2 for marine images. We encourage you to apply for both projects. 

Position 1: medical imaging 

You are keen on contributing to new advances in deep learning methodology for cardiac ultrasound imaging. Deep learning methods for cardiac ultrasound can e.g. be used for automatic measurements from image sequences, to assess im-age quality, or take benefits of recent research results in foundational neural models, where models learn from large unlabelled image datasets, but also on additional data like clinical reports or electronic health records. The work will be done in collaboration with GE Healthcare and their research center in Oslo. 

You will focus on challenges related to self-supervised learning, and combine this with either modelling knowledge, incorporate reliability/uncertainty, and/or explainable models. 

Position 2: marine imaging

You are keen on contributing to new advances in deep learning methodology for monitoring marine ecosystems. This can be data collected about the marine environment from e.g., satellites, sonar images, or underwater video from AUVs. The research will be done in cooperation with the Institute of Marine Research. 

The core of the position will be development of new deep learning methods for segmentation/classification of data with limited and weak labels and imbalanced data sets. You can also focus on challenges related to modelling prior knowledge, incorporate uncertainty, or methods with improved explainability. 

For more information and how to apply: https://www.jobbnorge.no/en/available-jobs/job/257213/phd-research-fellow-in-deep-learning-on-image-data