MR Image Synthesis

Supervision by: Prof. Christian Wachinger
Contact: wachinge@in.tum.de

Project Description

Magnetic resonance imaging (MRI) enables the acquisition of images with various contrast to display different properties of the underlying anatomy. In neuroimaging, T1-weighted and T2-weighted images are among the most popular imaging contrasts. While acquiring multiple scans with different contrasts of a patient can help in the diagnosis, it also leads to a dramatic increase in scanning time. In this project, we will investigate if it is possible to synthesize one image contrast from the other. Answering the question whether it is possible to create a T1-weighted MRI scan from a T2-weighted scan, without actually acquiring the T1 scan. The ability to synthesize high quality scans can have wide clinical impact. In terms of methods, we will investigate latest machine learning techniques to address this interesting problem.

Requirements

• We are looking for an enthusiastic student with programming skills
• Experience in Image Processing and Machine Learning is a plus.

Contact

If you are interested, please contact Christian Wachinger.