Vacancy for scientific programmer for an online competition platform and cloud-based deployment of medical image analysis systems

The Diagnostic Image Analysis Group at the Department of Radiology, Radboud University Medical Centre, is offering a position for a scientific programmer.

Organization

The Diagnostic Image Analysis Group (DIAG) is a research division of the Department of Radiology and Nuclear Medicine of the Radboud University Medical Center. We develop, validate and deploy novel medical image analysis methods, usually based on machine learning technology and focusing on computer-aided diagnosis. Application areas include lung imaging, breast imaging, prostate imaging, and brain imaging. We closely collaborate with clinicians and industry.

Background

Many groups worldwide, both from universities and companies, are developing computer systems to analyze medical data. Often multiple approaches for one particular task, for example the detection of lung cancer in chest CT scans, have been described in scientific publications or are commercially available. It has become popular to invite developers of such systems to evaluate their method on a common data set, in so called challenges or competitions, to provide a fair comparison and investigate if a better system can be constructed by combining multiple approaches.

DIAG has pioneered the organization of these challenges, developed COMIC, an open source library to host challenges, and we maintain an overview of medical image challenges on our grand challenge web site.

We have started a collaboration with Microsoft Research who developed CodaLab, an open-source web-based platform that enables researchers, developers, and data scientists to collaborate and organize competitions. CodaLab is in many ways similar to COMIC, but has a broader focus. Both systems have been written in the Django Python web framework.

Job

This scientific programmer position is funded by Microsoft Research and the main aim is to integrate a number of features from COMIC in CodaLab and extend CodaLab so that it will be possible to run medical image analysis algorithms in the cloud, starting with the Microsoft Azure cloud platform. You should also use the developed functionality in several competitions we plan to organize and in ongoing research projects within DIAG.

You will work within the scientific programming team of DIAG led by Sven Lafebre and you will be supervised by him and Bram van Ginneken. You will actively collaborate with CodaLab developers from leading universities in the United States (Harvard Medical School, Stanford) and from Microsoft Research in Seattle, Washington. We also closely collaborate with Fraunhofer MEVIS in Bremen, Germany, the largest group in medical image analysis in Europe, and we collaborate with other medical imaging groups in Europe on COMIC.

Your profile

We are looking for a highly motivated and talented software developer with

  • Excellent programming skills
  • Experience in web technology, in particular Django and Python
  • Strong interest in medical image analysis
  • Excellent communication skills

In addition, experience in cloud services development, experience with C++, an active track record in open source development, and knowledge about certified software development are advantages.

Requirements

As a scientific programmer you should have an MSc or PhD degree in Computer Science, Physics, Engineering or Biomedical Sciences or similar. Salary will be in agreement with your experience. Employment will be initially for one year, with a possible extension.

Opportunities

We offer a challenging research environment in a growing international research group with ample opportunities to further develop your skills. If desired, you can spend part of your time on research.

Information

For more information contact Bram van Ginneken by e-mail.

Apply

To apply, use the button at the bottom of this page.