Robust and accurate cerebral hemisphere segmentation in non-contrast CT using a 2.5D Dense U-Net
Keywords
No Thumbnail Available
Authors
Issue Date
2018-06-28
Language
en
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This work is about automatically segmenting cerebral hemispheres using
a convolutional neural network. The aim is to publish the work in NeuroImage:
Clinical, since it is the rst work to automatically segment cerebral
hemisphere in non-contrast CT scans. Some sections can lack background
information, for the sake of conciseness of being it a paper for a targeted
journal. The background information that is relevant can be found in the
included appendices. Also more work that is done in the framework of the
extended research project is included in the appendices
For the last eight months I worked on this extended research project at
the Radboud Hospital Nijmegen in the Digital Image Analysis group (DIAG).
Besides the fact that I learned much in this period, be it academical, medical
or technical, I had a very good time here. I would like to thank this group
for the excellent guidance, especially Ajay and Rashindra, who supervised
me through the entire process.
I would also like to thank George for supervising on behalf of the university.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen