Prof. Dr. Susanne Schnell

AI in Magnetic Resonance Imaging

Prof. Dr. Susanne Schnell, Chair of Medical Physics, is developing new MRI measurement sequences as well as image processing and data analysis methods with a focus on vascular diseases (e.g. atherosclerosis) and cardiac MRI. AI methods are used in image processing and prognosis. In image processing, AI is used on the one hand to improve the data basis, for example by significantly improving the spatial resolution using AI, which enables more precise quantification of haemodynamic parameters and a reduction in MRI measurement time. On the other hand, AI methods are used in the segmentation of cerebral arteries in 4D blood flow measurement using MRI. This enables full automation of image analysis and quantification of hemodynamic parameters such as flow rate, pressure difference or wall shear stress. Susanne Schnell is also working on automated evaluations in MR cardiac imaging to quantify cardiac strain using AI and on prognoses for a second stroke in patients with intracranial atherosclerotic stenosis.

The following projects have been submitted or published as articles:

1. E. Ferdian, D. Marlevi, J. Schollenberger, M. Aristova, E.R. Edelman, S. Schnell, C.A. Figueroa, D.A. Nordsletten, A.A. Young. Cerebrovascular super-resolution 4D Flow MRI – using deep learning to non-invasively quantify velocity, flow, and relative pressure, Submitted to Medical Image Analysis

2. H Berhane, M Aristova, Y Ma, M Markl, and S Schnell, Fully Automated Intracranial Vessel Angiogram Segmentation from 4D flow MRI Data in Intracranial Stenosis Patients using Deep Learning, abstract 1665, Proceedings International Society of Magnetic Resonance in Medicine (ISMRM) 2020 Sydney