Medical Metrology

Quantitative MRI yields a wide range of (bio-)physical parameters which provide important diagnostic information about the human body. A major challenge is the accurate estimation of these parameters while keeping scan times to a minimum. Physics-informed deep-learning is a promising approach to solve this mathematically challenging, non-linear problem.

If you are studying physics, electrical engineering, computer science or a comparable area of engineering/natural sciences and you are interested in a summer project or BSc/MSc project in the field of motion compensation, please contact Christoph Kolbitsch (📧 christoph.kolbitsch(at)ptb.de).

Breathing or the beating of the heart can lead to movement of organs in the human body. This physiological motion can strongly impair the quality of MR images. We develop novel approaches to minimise these motion artefacts and to ensure excellent diagnostic accuracy of MRI for a wide range of applications, such as quantitative cardiac MRI or PET-MR.

If you are studying physics, electrical engineering, computer science or a comparable area of engineering/natural sciences and you are interested in a summer project or BSc/MSc project in the field of motion compensation, please contact Christoph Kolbitsch (📧 christoph.kolbitsch(at)ptb.de).

In our research group, we are developing new methods for magnetic resonance imaging (MRI). Our focus is put on flow imaging, which enables the time-resolved and non-invasive imaging of flow profiles in blood vessels. In its clinical application, severity of pathologies can be classified without catheterization. In addition, turbulent flow encoding allows for improved decision decision-making while treating aortic valve diseases.

However, the MRI method has intrinsic errors which must be characterized and corrected. Of special interest are phantoms (= artificial object to be scanned in the MRI) which can range from fluid-filled spheres to artificial heart depending on their complexity. We at PTB strive for highest accuracy and repeatability – to this end we want to improve a motion phantom ourselves and make its hardware and software available open-source. With this, an important milestone in the comparability of MR centers and, at the end of the line, best patient care could be provided.

In this work a control for an existing rotational motion phantom [1] shall be designed. The tasks comprise

• system identification, control design, simulation
• set-up of the phantom and (guided) imaging at the MRI scanner
• evaluation of the acquired MRI data
• documentation and open-source publication of the project

We seek master students

• with a major in electrical engineering, mechanical engineering, physics or similar;
• with knowledge about control engineering;
• having good programming skills (preferably Python und MATLAB) and
• with keen interest in MRI

If you are interested, please contact Hannes Dillinger or Sebastian Schmitter and please also send your CV and scores:

Dr. Hannes Dillinger ( 📧 hannes.dillinger(at)ptb.de , ☎ +49 30 3481-7974 )
Dr. Sebastian Schmitter ( 📧 sebastian.schmitter(at)ptb.de , ☎ +49 30 3481-7767 )

We look forward to hearing from you.

https://www.ptb.de/cms/ptb/fachabteilungen/abt8/fb-81/ag-814.html

[1] Vali, A, Schmitter, S, Ma, L, et al. Development of a rotation phantom for phase contrast MRI sequence validation and quality control. Magn Reson Med. 2020; 84: 3333-3341. doi: 10.1002/mrm.28343