Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Deep Learning Enabled Image Registration for Magnetic Resonance and Optoacoustic Tomography Data
This project aims to combine deep learning based image segmentation algorithm with an automated registration framework for optoacoustic and magnetic resonance images.
Optoacoustic (OA) imaging is a new emerging technique based on the illumination of an object with different wavelengths and reconstruction of resulting signals with backprojection and model based algorithms. Magnetic resonance imaging (MRI) and OA imaging are two complementary modalities for quantification in pathophysiology. Accurate registration of mouse MRI and OA images helps scientists to understand molecular events that occur during the development of Alzheimer’s disease. This project aims to create a deep learning enabled framework for the registration of MRI and OA images that are acquired from mouse models of Alzheimer’s disease.
Optoacoustic (OA) imaging is a new emerging technique based on the illumination of an object with different wavelengths and reconstruction of resulting signals with backprojection and model based algorithms. Magnetic resonance imaging (MRI) and OA imaging are two complementary modalities for quantification in pathophysiology. Accurate registration of mouse MRI and OA images helps scientists to understand molecular events that occur during the development of Alzheimer’s disease. This project aims to create a deep learning enabled framework for the registration of MRI and OA images that are acquired from mouse models of Alzheimer’s disease.
At the first step of this project, the student is expected to combine already implemented OA image segmentation algorithm (U-Net [1]) with registration framework that is implemented by our group in the following paper [2]. The second step includes the improvement of segmentation and registration algorithms by different deep learning based methods. The student is expected to work independently and come up with new solutions for the second part of the project.
[1] Olaf Ronneberger, Philipp Fischer, Thomas Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation”, Medical Image Computing and Computer-Assisted Intervention (MICCAI), Springer, 2015.
[2] Wuwei Ren, Hlynur Skulason, Felix Schlegel, Markus Rudin, Jan Klohs, Ruiqing Ni, “Automated registration of magnetic resonance imaging and optoacoustic tomography data for experimental studies”, Neurophotonics 2019.
At the first step of this project, the student is expected to combine already implemented OA image segmentation algorithm (U-Net [1]) with registration framework that is implemented by our group in the following paper [2]. The second step includes the improvement of segmentation and registration algorithms by different deep learning based methods. The student is expected to work independently and come up with new solutions for the second part of the project.
[1] Olaf Ronneberger, Philipp Fischer, Thomas Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation”, Medical Image Computing and Computer-Assisted Intervention (MICCAI), Springer, 2015.
[2] Wuwei Ren, Hlynur Skulason, Felix Schlegel, Markus Rudin, Jan Klohs, Ruiqing Ni, “Automated registration of magnetic resonance imaging and optoacoustic tomography data for experimental studies”, Neurophotonics 2019.