A multicenter study to implement an imaging biomarker for acute stroke management: deep learning-based quantification of net water uptake in the ischemic brain

• Applicants: Dr. Gabriel Broocks; Privatdozent Dr. René Werner
• Subject Area: Clinical Neurology; Neurosurgery and Neuroradiology
• Term: since 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 514830458

In ischemic stroke, the indication for thrombolytic and endovascular treatment depends on the time from symptom onset to imaging. In cases of unknown symptom onset or in the extended time window, neuroimaging is required to guide treatment decisions, such as magnetic resonance imaging (MRI), or computed tomography (CT) perfusion. The utilization of CT in the acute situation has the advantage of higher availability, applicability, and speed. However, to indicate the application of intravenous alteplase (i.e. EXTENT trial) or endovascular treatment (i.e. DEFUSE-3 trial), commercial CT perfusion analysis tools are required as stated in the current guidelines. This project aims to develop and validate a deep learning solution for automated CT-based lesion water uptake quantification as feasible alternative to identify patients within a suitable time window for intravenous thrombolysis.

• DFG Programme: Research Grants
• International Connection: USA
• Co-Investigators: Professor Dr. Jens Fiehler; Dr. Jan Erik Gewehr; Privatdozentin Dr. Uta Hanning; Professor Dr. André Kemmling; Professor Dr. Sönke Langner
• Cooperation Partner: Professor Dr. Michael H. Lev