Medical Image Computing

Module aims

The course covers the fundamental concepts and methodologies of medical image computing and image analysis and relates those to clinical applications in diagnosis, therapy and intervention. The aim is to provide an overview of the different areas, such as image processing, registration and segmentation, with an emphasis on understanding the theoretical and practical aspects of various methods. The necessary skills will be taught that enable students to work and conduct research in medical image computing.

Learning outcomes

Knowledge and Understanding

To know the specific material covered in the Syllabus, including the ability to do the following:

To acquire a basic understanding of the important theoretical concepts and practical skills related to medical image computing and image analysis

To understand the limitations of medical image analysis techniques

Intellectual Skills

To compare, characterize and evaluate different approaches for practical problems

To design medical image analysis systems using the basic techniques

To specify which algorithms can be applied to which problems and image modalities

Practical Skills

To apply techniques for image enhancement, segmentation and registration of 2D, 3D, and 4D medical images

To extract clinically useful information from images using image computing algorithms

To implement medical image processing pipelines that solve clinical image analysis tasks

Module syllabus

The course covers the following topics. Within each topic fundamental algorithms will be discussed:

  • Introduction to Medical Image Computing and Toolkits
  • Medical Image Visualization (Multi-Planar Rendering, Volume Rendering, Surface and Mesh Rendering)
  • Medical Image Analysis (Filtering in Spatial and Frequency Domain, Anisotropic Diffusion)
  • Medical Image Registration (Point-based and Intensity-based Registration)
  • Medical Image Segmentation (Clustering, Region-based and Model-based Methods, Graph Cuts, Markov Random Fields)
  • Machine Learning in Medical Image Analysis (Dimensionality Reduction, Supervised and Unsupervised Classification/Regression)


None. Having done courses CO316 Computer Vision and CO395 Machine Learning is beneficial.

Teaching methods

A total of 8 sessions

Sessions are alternating weekly between 2 hours lectures, and 1.5 hours tutorials

Tutorials are informal and interactive; each comes with detailed questions and answers


*This is a level 7/M course

One assessed coursework involving both analytical and programming tasks (typically submitted at the end of the term) and an exam (typically choose 2 out of 3 questions)

Reading list


Module leaders

Dr Ben Glocker