This repository contains my work for the clinical project (Klinisches Anwendungsprojekt) which is part of my Bachelor program at the Technical University of Munich (TUM).
The primary goal of this project is to predict MRI-based lesion masks from CT data, effectively bridging the gap between acute CT imaging and follow-up MRI. This has important clinical implications for early stroke assessment and treatment planning.
report/: Contains the LaTeX source files for the project reportmain.pdf: The current version of the compiled report
development/: Contains scripts used during development
The analysis is based on the ISLES24 dataset, which includes:
- CT perfusion images
- Derived perfusion maps (CBF, CBV, MTT, Tmax)
- MRI-derived lesion masks
Felician Schwarz
This project is part of academic work at TUM and is not licensed for public use without permission.