Course syllabus for Medical image processing

Course syllabus adopted 2022-02-17 by Head of Programme (or corresponding).

Overview

  • Swedish nameMedicinsk bildbehandling
  • CodeSEE120
  • Credits7.5 Credits
  • OwnerTKMED
  • Education cycleFirst-cycle
  • Main field of studyBiomedical engineering
  • DepartmentSPACE, EARTH AND ENVIRONMENT
  • GradingTH - Pass with distinction (5), Pass with credit (4), Pass (3), Fail

Course round 1

  • Teaching language English
  • Application code 73118
  • Maximum participants80
  • Open for exchange studentsNo
  • Only students with the course round in the programme overview.

Credit distribution

0121 Project 1.5 c
Grading: UG		 1.5 c
0221 Examination 6 c
Grading: TH		 6 c
  • 02 Jun 2023 pm J
  • 04 Jan 2023 am J
  • 22 Aug 2023 pm J

In programmes

Examiner

Go to coursepage (Opens in new tab)

Eligibility

General entry requirements for bachelor's level (first cycle)
Applicants enrolled in a programme at Chalmers where the course is included in the study programme are exempted from fulfilling the requirements above.

Specific entry requirements

The same as for the programme that owns the course.
Applicants enrolled in a programme at Chalmers where the course is included in the study programme are exempted from fulfilling the requirements above.

Aim

This course uses examples with medical image processing to teach fundamental knowledge about two and three-dimensional signal processing while giving students a chance to develop their programming skills via project work. More specifically the course teaches the main techniques of Image Processing needed to prepare medical (and other) images for human interpretation or subsequent automated image analysis. These methods respectively improve subjective image quality  (image enhancement), remove known image distortions such as blurring effects (image restoration), reduce image data sizes for storage or transmission (image compression) or form images from indirectly sampled data; such as from projections (image reconstruction). On completion of the course  students should be able to implement simple customized versions of the major image processing algorithms used in medical image processing via coding in MATLAB. Although medical applications of image processing will be emphasized, and most examples will be taken from medicine,  some applications of the techniques in other fields will also be presented.

Learning outcomes (after completion of the course the student should be able to)

• Visualise via means of mental images the process of forming 1D and 2D Fourier transforms and the convolution process. Be able to quantify and explain to others the  practical effects on imaging of using the Discrete Fourier Transform (aliasing etc) and be able to implement methods for eliminating such effects (image padding etc).

• Apply knowledge about the human vision system to implement image enhancement methods for human end use. Choose and apply appropriate image enhancement methods for different applications. Discriminate between cases where automated image enhancement methods produce appropriate results and where they do not.

• Choose appropriately between averaging and median filtering for reducing image noise based on noise statistics.

• Code and apply image smoothing and sharpening techniques to images using both image and Fourier domains methods. Be able to select between optimum methods of edge detection for different applications.

• Compute manually the convolution of matrices representing images with point spread functions (PSFs).  Estimate the tradeoff  between improved image sharpness and increased noise on applying different image restoration algorithms and appropriately choose which method to apply in specific cases.

• Explain to others the nature of wavelets and be able to apply wavelets to de-noise images.

• List common image formats used in medicine.  Be able to justify the use of lossless versus lossy compression for different applications.

• Describe and implement in software simple methods of image registration.

• Code methods of image reconstruction from projections as used in  X-ray and PET Computed Tomography. Be able to explain  the nature of residual image artifacts and propose methods for their removal.

• Implement simple methods of Magnetic Resonance Imaging from observed data.

Content

Brief introduction to medical imaging modalities (Computed Tomography (CT) X-ray, Positron Emiision Tomography (PET), Magnetic Resonance Imaging (MRI), ultrasound, optical imaging). Basics of the human vision system. Image Enhancement: transform functions and histogram equalisation; image smoothing and sharpening in 2D and 3D; edge detection and noise reduction. Simple segmentation methods. Fourier domain methods of image enhancement and the implementation of such methods via 2D Discrete Fourier Transforms.  Basic introduction to wavelets and their applications to image de-noising.  Medical Image formats. Image Registration methods. The difference between lossy and lossless compression. Lossless Image Compression as implemented by Huffman coding, multi-pixel coding and run length coding. General methods of image restoration including inverse and pseudoinverse filters and the Wiener filter. Image reconstruction from projections including filtered back-projection and algebraic methods. Basic methods for MRI reconstruction.

Organisation

Lectures, lab exercises, problem classes and project.

Literature

Digital Image Processing - 4th Edition - Global edition-  by Gonzalez and Woods.

Examination including compulsory elements

Compulsory Project and graded written exam.

The course examiner may assess individual students in other ways than what is stated above if there are special reasons for doing so, for example if a student has a decision from Chalmers on educational support due to disability.