Course content

The course teaches key principles and methods for processing and automated classification of digital images, with focus on medical images. Central topics in the course are normalization, filtering, segmentation, texture analysis, fusing and automated classification. The most important syntax in programming and what programs and algorithms are best suited for various medical modalities and research questions will be thought throughout the course.

Learning outcome

A candidate that has completed the course should have the following learning outcomes: The candidate: Knowledge: - Has advanced knowledge of the structure and properties of digital images - Has in-depth knowledge of the principles of normalization, filtering and other basic methods of image processing - Can evaluate possibilities and limitations in different types of image processing (segmentation, visualization, texture analysis and fusing) - Can use knowledge to assess new research in imaging and automated classification Skills: - Can use relevant imaging tools independently - Can perform various types of image processing (segmentation, visualization, texture analysis and fusing) - Can carry out an independent and refined project based on problem and untreated images General competence: - Can discuss issues, analyzes and conclusions within image processing with collaborative professional groups, colleagues and students - can analyze issues and results in research with image processing and automated classification - Masters the terminology of image processing and central programming syntax

Learning methods and activities

Blockwise lectures on campus and group work with electronic submission. There will be sessions over up to a total of 6 days during the semester, and the students have 4 compulsory assignments. The work requirements will be student-active, with exercises with image data; either pictures taken at their own workplace, or customized data sets. The assignments can also be linked to ongoing research projects at the department. All assignments come with written and / or video-based tutorials. Work requirements facilitate thematic interconnection of theory and practice. All learning resources are made available via the e-learning platform: articles, links, short films (animations, lectures, etc.), self-tests, answers to assignments and discussion forums with students and teachers.

Specific conditions

Required previous knowledge

The course is reserved for students enrolled to the Master of Science in Medical Imaging Technologies.

Course materials

Gonzalez and Wood: Digital Image Processing. 4th Edition, 2018, Pearson, NY, USA