Enhancement of clinical optoacoustic and ultrasound images
- Sponsored by: Institute for Biological and Medical Imaging, Helmholtz Zentrum München
- Scientific Lead: Dr. Dominik Jüstel
- Co-Mentor: PhD candidate Michael Rauchensteiner
- Project Lead: Dr. Ricardo Acevedo Cabra
- Term: Winter semester 2018
Modern machine learning frameworks that are trained by high-quality data have proven to be superior to many classic methods in image processing in terms of accuracy and efficiency. A particular application is the enhancement of medical images by learning high-quality reconstructions from fast low-quality reconstructions. This method is an instance of image-to-image translation and has the potential to achieve high quality imaging in real time.
Explore the applications of these novel methods for a state-of-the-art integrated optoacoustic and ultrasound imaging system (OPUS) at the interdisciplinary group of the Institute for Biological and Medical Imaging (IBMI) at the Helmholtz Center Munich and the Chair for Biological Imaging (CBI) at TUM, located at TranslaTUM next to the Klinikum rechts der Isar.
The goal of this project is to develop a method that enhances real-time ultrasound and/or optoacoustic images to improve the usability of the system in the clinic and to simplify data analysis. The particular tasks are: ensuring data quality by selection and augmentation, design and training of a suitable architecture, and evaluation of the developed method.
Results: The results of this project are explained in detail in the final documentation and presentation.