This title appears in the Scientific Report :
2013
Please use the identifier:
http://hdl.handle.net/2128/5401 in citations.
GPU-accelerated Segmentation of high-resolution Human Brain Images acquired with Polarized Light Imaging
GPU-accelerated Segmentation of high-resolution Human Brain Images acquired with Polarized Light Imaging
High-resolution three-dimensional polarized light imaging (PLI) is an approach pursued by the Institute of Neuroscience and Medicine at Forschungszentrum Jülich to map nerve fibers and their pathways in human brains. Sections of cut post-mortem brains are imaged with a microscopic device. A section...
Saved in:
Personal Name(s): | Westhoff, Anna Maria (Corresponding author) |
---|---|
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Imprint: |
Jülich
Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag
2013
|
Physical Description: |
84 p. |
Dissertation Note: |
FH Aachen-Jülich, Masterarbeit, 2013 |
Document Type: |
Master Thesis Internal Report |
Research Program: |
Computational Science and Mathematical Methods |
Series Title: |
Berichte des Forschungszentrums Jülich
4365 |
Subject (ZB): | |
Link: |
OpenAccess |
Publikationsportal JuSER |
High-resolution three-dimensional polarized light imaging (PLI) is an approach pursued by the Institute of Neuroscience and Medicine at Forschungszentrum Jülich to map nerve fibers and their pathways in human brains. Sections of cut post-mortem brains are imaged with a microscopic device. A section is moved during the imagingwithin the microscope so that a mosaic of about 30x30 image tiles is created for a gross histological human brain section. These up to 900 tiles per section and about 1500 sections in total have to be handled in the 3D reconstruction of the brain. A way to accelerate this process is a previous segmentation of the image tiles which leads to black-and-white masks marking the brain and background pixels of the original tiles. Hence, all non-brain parts of the tiles can be ignored during the reconstruction.A region growing segmentation is developed and implemented for the PLI data. The challenge to adapt this algorithm to the given dataset is to automatize the choice of seeds needed as starting points for the growing process. Therefore, an automated method of seed determination has to be developed. It uses statistics of the whole brain based on the joint intensity histogram. This approach leads to a minimal fixed amount of required manual input which is independent of the number of image tiles to be segmented. The software is parallelized for the GPU cluster JUDGE, i.e. it combines two levels of parallelism, namely a multicore implementation and the data parallel execution of appropriate subtasks on a GPU. This leads to a well-scaling application that achieves the expected segmentation results. |