|
Aim of the course:
- After attending the course, students will have a basic knowledge and understanding of image processing, i.e. linear filtering and morphological operations. Furthermore, they will have an overview and basic understanding of a variety of principles for image segmentation. They will also understand principles of 3D computer graphics and visualization.
- Students will gain the skills to implement some of these techniques in Matlab, and will be exposed to basic Python. This will equip them with the tools and skills for actual development of systems and for conducting experiments during their training in M2 and M3.
- Students will be able to recognize and evaluate the applicability of these techniques in clinical problems related to diagnosis and interventions. If a clinical application demands it, students will be able to analyze the situation in terms of a mathematical model, which will then form the basis for an algorithm.
|
|
The course describes the necessary trajectory for the quantitative analysis, the segmentation and the visualisation of medical imaging data. The following steps can be distinguished:
- Image formation.
- Image acquisition.
- Reconstruction from projections.
- Quantitative image analysis.
- Image segmentation.
- Imaging registration.
- 3D visualisation.
The first three steps provide the basic knowledge about medical imaging systems. These have been treated in the course ”Imaging techniques" and will not be repeated here. In the current course, imaging is approached at a system level using a linear description of image formation as a central starting point.
The focus of the course is on steps 4 – 7. The end result applies to diagnostics and interventions. The theory of the course is presented in lectures; Throughout the course, four image processing labs will provide exercise and experience in the application of the theoretical concepts to medical image data. These labs will be graded for each student. Short workshops will enable the student to master the abstract mathematical descriptions of the image processing algorithms.
The course ends with a project concerning image data from the medical (TM) practice. The project will be assigned from a pool of available projects randomly to pairs of students. Endpoint will include a written report and a in-class presentation. There is no final written exam (or a resit) for this course. The final grade will be based on assignments (15% each, total of 60%), and implementation (15%), report delivery (10%), and presentation (15%) of the project.
This course is only open for TM students. BME students who are interested in image analysis and segmentation are advised to attend "Image Processing and Computer Vision" (191210910) as an optional course.
|
 |
|