PhD position geometric deep learning
Wednesday 22nd July 2020
Closing Date: August 23, 2020
Medical imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound are at the heart of clinical diagnostics. In recent years, artificial intelligence techniques, and especially deep learning, have already pushed the state-of-the-art in many tasks including 3D medical image segmentation. Automatically obtained segmentations can be further used as input for disease classification, quantification, or for the extraction of functional information. Such downstream applications require efficient methods that operate on 3D shapes. In this position, you will develop machine learning techniques to bridge the gap between 3D model extraction from medical images and downstream applications. This requires the development of new geometric deep learning approaches that operate on 3D shape representations such as triangular meshes. You will work at the intersection of mathematics, computer science, artificial intelligence, and physics. The methods developed in this project will be used to tackle real-world clinical problems.
We look for a highly motivated, enthusiastic researcher who is driven by curiosity and has:
- or will shortly acquire, a M.Sc. degree in (applied) mathematics, computer science, biomedical engineering, (applied) physics, artificial intelligence or a related field.
- affinity with the clinical application of artificial intelligence techniques.
- a creative and interdisciplinary approach to pushing boundaries.
- programming experience (e.g. C++, Matlab, Python, TensorFlow, PyTorch).
- experience with deep learning/convolutional neural networks.
- proficiency in English
For more information you are welcome to contact Dr. Jelmer Wolterink (email@example.com).
|Organization||University of Twente|
|Location||Enschede, The Netherlands|
|Title||PhD position geometric deep learning|