MSc+PhD vacancy for body composition assessment in 3D CT and MR images

MSc+PhD vacancy for body composition assessment in 3D CT and MR images

Clinical problem

The composition of the body in terms of muscle and fat mass is a precise measure of the physical state of the patient. This information has many practical applications in clinical routine care, such as monitoring treatment response to chemotherapy and immunotherapy in patients with cancer, or predicting clinical outcomes in other patients. While every hospital routinely acquires hundreds of cross-sectional imaging studies (CT and MRI scans) every day that would be suitable for deriving precise body composition measurements, this potential currently remains unused. Measuring fat and muscle mass would require manually delineating tissues depicted in the scan, which is labor-intensive and time-consuming.

Aim of the project

Your task is to develop deep learning algorithms for automatic delineation (segmentation) of fat and muscle compartments in 3D CT and MRI scans. These segmentations will then be used to allow radiologists to routinely include fat and muscle mass into their reports. Next to semantic segmentation in medical images, the project also involves finding solutions for various practical challenges, such as efficient data annotation, automatic recognition of suitable scans, and extrapolation in scans with incompletely depicted body parts.

The project offers the opportunity to work with multiple types of commonly acquired medical images, with multiple hospitals and an industrial partner, and to develop software that would immediately provide benefits to radiologists, clinical researchers and other clinicians.

Combining MSc and PhD degree

In collaboration with Siemens Healthineers and the Jeroen Bosch Ziekenhuis, this project provides the opportunity to continue your research beyond a MSc thesis and also obtain a PhD degree. The project consists of a 6-months research project at the Diagnostic Image Analysis Group at Radboudumc, followed by a 6-months international internship at Siemens Healthineers' campus in Princeton in the United States (both are paid internships). After defending your MSc thesis, you continue the project at Radboudumc as PhD candidate with the aim to obtain the PhD degree within 3 years.

Requirements

  • Student in the final year of a Master study in biomedical engineering or technical medicine, computer science, data science, artificial intelligence or a related area.
  • Good programming skills (Python) and a theoretical understanding of deep learning architectures are important.
  • Experience with medical images and medical image analysis is a plus.

Applying for this project

For more information, please contact Matthieu Rutten or Nikolas Lessmann.

To apply for this project, please e-mail your CV, list of grades, and a motivation letter with a short summary of your background to Nikolas Lessmann. We process applications immediately until the position is filled.

People

Matthieu Rutten

Matthieu Rutten

Associate Professor, Radiologist

Diagnostic Image Analysis Group

Nikolas Lessmann

Nikolas Lessmann

Assistant Professor

Diagnostic Image Analysis Group

Alina Vrieling

Alina Vrieling

Assistant professor

Health Evidence, Radboudumc

Bram van Ginneken

Bram van Ginneken

Professor

Diagnostic Image Analysis Group