Methods to Automatically Transform MR Images of the Brain into the AC-PC Coordinate System
(2024) In Master's Theses in Mathematical Sciences FMAM05 20241Mathematics (Faculty of Engineering)
- Abstract
- One way to reduce the growing workload for radiologists is to automate the trans-
formation of 3D MR images of the brain into the AC-PC coordinate system. In this
thesis, both non-deformable and deformable image registration methods were used
to register images to a template in the desired space, the AC-PC coordinate system.
First, three non-deformable transformations were used together with three different
templates and two similarity metrics. To transform an image to the AC-PC coordin-
ate system, only the rotation matrix and translation vector are needed. Thus, the
transformation matrix and the landmarks (AC, PC and MS) were used to extract the
rotation matrix and translation vector from non-deformable and deformable image re-
... (More) - One way to reduce the growing workload for radiologists is to automate the trans-
formation of 3D MR images of the brain into the AC-PC coordinate system. In this
thesis, both non-deformable and deformable image registration methods were used
to register images to a template in the desired space, the AC-PC coordinate system.
First, three non-deformable transformations were used together with three different
templates and two similarity metrics. To transform an image to the AC-PC coordin-
ate system, only the rotation matrix and translation vector are needed. Thus, the
transformation matrix and the landmarks (AC, PC and MS) were used to extract the
rotation matrix and translation vector from non-deformable and deformable image re-
gistration, respectively. The combinations of transformation, template and similarity
metric were evaluated with four different metrics: distance to AC, angle in sagittal,
coronal, and axial plane. The non-deformable transformations were rigid, similarity
and affine transformation, while the templates were Colin27 T1- and T2-weight, and
MNI305. The two similarity metrics used were Mattes mutual information, which is a
modification to mutual information, and cross correlation. The best combination in the
non-deformable case was the affine transform with Colin27 with T1-weight and Mattes
mutual information. The average angular errors for this combination were 3.45°, 0.87°
and 1.18° for the angles in the sagittal, coronal, and axial plane, respectively. The
average distance error to AC was 3.11mm. The second part was the development of a
deformable transformation which was semi-based on the first part. This because it used
the best template and the best similarity metric from the first part. The deformable
transformation was symmetric normalization (SyN) with Colin27 with T1-weight and
Mattes mutual information. The average angular errors for the deformable method
were 1.84°, 1.24° and 1.12° for the angles in the sagittal, coronal, and axial plane,
respectively. The average distance error to AC was 1.47mm. The average time for a
non-deformable registration was 18 seconds, and 320 seconds for a deformable regis-
tration. Except for the time, the deformable method was on par with state-of-the-art
machine learning methods from the literature. The non-deformable method performed
worse than the methods from the literature, except when comparing the time for one
registration. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9160025
- author
- Larsson, Malin LU
- supervisor
- organization
- alternative title
- Metoder för att automatiskt transformera MR bilder av hjärnan till AC-PC koordinatsystemet
- course
- FMAM05 20241
- year
- 2024
- type
- H2 - Master's Degree (Two Years)
- subject
- publication/series
- Master's Theses in Mathematical Sciences
- report number
- LUTFMA-3533-2024
- ISSN
- 1404-6342
- other publication id
- 2024:E20
- language
- English
- id
- 9160025
- date added to LUP
- 2024-06-28 16:00:17
- date last changed
- 2024-06-28 16:00:17
@misc{9160025, abstract = {{One way to reduce the growing workload for radiologists is to automate the trans- formation of 3D MR images of the brain into the AC-PC coordinate system. In this thesis, both non-deformable and deformable image registration methods were used to register images to a template in the desired space, the AC-PC coordinate system. First, three non-deformable transformations were used together with three different templates and two similarity metrics. To transform an image to the AC-PC coordin- ate system, only the rotation matrix and translation vector are needed. Thus, the transformation matrix and the landmarks (AC, PC and MS) were used to extract the rotation matrix and translation vector from non-deformable and deformable image re- gistration, respectively. The combinations of transformation, template and similarity metric were evaluated with four different metrics: distance to AC, angle in sagittal, coronal, and axial plane. The non-deformable transformations were rigid, similarity and affine transformation, while the templates were Colin27 T1- and T2-weight, and MNI305. The two similarity metrics used were Mattes mutual information, which is a modification to mutual information, and cross correlation. The best combination in the non-deformable case was the affine transform with Colin27 with T1-weight and Mattes mutual information. The average angular errors for this combination were 3.45°, 0.87° and 1.18° for the angles in the sagittal, coronal, and axial plane, respectively. The average distance error to AC was 3.11mm. The second part was the development of a deformable transformation which was semi-based on the first part. This because it used the best template and the best similarity metric from the first part. The deformable transformation was symmetric normalization (SyN) with Colin27 with T1-weight and Mattes mutual information. The average angular errors for the deformable method were 1.84°, 1.24° and 1.12° for the angles in the sagittal, coronal, and axial plane, respectively. The average distance error to AC was 1.47mm. The average time for a non-deformable registration was 18 seconds, and 320 seconds for a deformable regis- tration. Except for the time, the deformable method was on par with state-of-the-art machine learning methods from the literature. The non-deformable method performed worse than the methods from the literature, except when comparing the time for one registration.}}, author = {{Larsson, Malin}}, issn = {{1404-6342}}, language = {{eng}}, note = {{Student Paper}}, series = {{Master's Theses in Mathematical Sciences}}, title = {{Methods to Automatically Transform MR Images of the Brain into the AC-PC Coordinate System}}, year = {{2024}}, }