One million men receive a prostate cancer diagnosis, and three hundred thousand die from prostate cancer each year worldwide. Magnetic Resonance Imaging (MRI) represents a major breakthrough by accurately detecting clinically significant prostate cancers at an early, potentially more curable stage. The same MRI can also be used for better treatment targeting and for avoiding unnecessary systematic biopsies. MRI allows patient-specific guidance during surgery. Radboudumc has recently set up an MRI in its Medical Innovation and Technology expert Center. Interpreting MRI imaging during surgery is however very challenging and time-consuming. Technology for real-time accurate modeling is not available. Artificial Intelligence and more specifically Deep Learning has shown promising results segmentation of medical images, mainly for diagnostic use. For interventional purposes, speed and accuracy are still major challenges.
This project is part of a recently awarded EDL perspectief project. EDL aims to make deep learning much more efficient. This Ph.D. project aims to research deep learning that will demonstrably help clinicians to provide real-time modeling of MRI during an intervention. The research includes automatic segmentation using a massive MRI database. We closely work with clinicians in all projects and are collecting extensive data sets of expert annotations. State-of-the-art deep learning models will be continuously evaluated and implemented in a clinical prototype for validation and feedback. Results will be presented at scientific meetings and published in journals.
Radboudumc is a clinical expert on prostate MRI and technical expert in the field of prostate AI technology for over 20 years and an early adopter of deep learning in medical imaging. Deep learning is currently the most active research area within machine learning and computer vision, and medical image analysis. DIAG currently has 40 deep learning researchers focused on various medical image analysis topics. A team of scientific programmers is supporting our deep learning research, maintaining a large cluster of computers equipped with high-end GPUs for large-scale experimentation.
You will be part of a multidisciplinary team, consisting of machine learning and clinical researchers and collaborate with external partners. The main external partner for this project is Siemens Healthineers, but you will have opportunities to work with the other partners in the EDL project: AIIR Innovations, ASTRON, CWI, Cyclomedia, Cygnify, Donders Institute, FEI, 2getthere, GN Hearing, Holst Centre, ING, Intel, Irdeto, Lely, Mobiquity, NLeSC, NXP, NVIDIA, Océ, Schiphol, Scyfer, Sectra, Semiotic Labs, Sightcorp, Sorama, SURFsara, TASS International, Tata Steel, TU Dresden, Delft University of Technology, Eindhoven University of Technology, Thales, TNO, TomTom, University of Twente, University of Amsterdam, 3DUniversum, VicarVision, ViNotion, VU Amsterdam, Wageningen University & Research.
You are a creative and ambitious researcher with an MSc/Ph.D. degree in Computer Science, Data Science, Physics, Engineering or Biomedical Sciences or similar, with a clear interest in machine learning, deep learning, medical image analysis. Good communication skills and expertise in software development, preferably in Python/C++, are essential. Experience with deep learning should be evident from the (online) courses you've followed, your publications, GitHub account, etc.
Positions can be filled by either a Ph.D. student or postdoc.
The Diagnostic Image Analysis Group (DIAG) is a research division of the Department of Radiology and Nuclear Medicine of the Radboud University Medical Center Nijmegen. Nijmegen is the oldest Dutch city with a rich history and one of the liveliest city centers in the Netherlands. Radboud University has over 17,000 students. Radboud UMC is a leading academic center for medical science, education, and healthcare with over 8,500 staff and 3,000 students.
The focus of DIAG is the development and validation of novel methods in a broad range of medical imaging applications. Research topics include image analysis, image segmentation, machine learning, and the design of decision support systems. Application areas include neuro, breast, prostate, lung and retina imaging and digital pathology. Key to the success of the group is in the close cooperation with clinicians. Currently, the group consists of around 40 researchers. You may read our deep learning survey paper to get an idea of our work in this field.
For further information contact Henkjan Huisman