CMPT 880

Instructor

• Dr. Ghassan Hamarneh

Summary

• Topics in image analysis (segmentation, registration)
• Advanced techniques (deformable/physics-based models, graph cuts)

Presentations

• Comparison of Salient Point Detection Methods for 3D Medical Images (Oct 14, 2005)
  • cs880-salient-slides.pdf
• Graph-based Segmentation (Oct 28, 2005)
  • graph-segment.ppt
• Mixture Model Analysis of DNA Microarray Images (Oct 29, 2005)
  • cs880-gaussian-slides.pdf
• Cell Segmentation, Tracking, and Mitosis Detection Using Temporal Context (Nov 18, 2005)
  • cs880-cell-slides.pdf
• Cross Entropy: A New Solver for Markov Random Field Modeling and Applications to Medical Image Segmentation (Nov 25, 2005)
  • cs880-ce-slides.pdf

Project

• registration of n-dimensional images using SIFT feature

SFU Course Description

SFU Computing Science 05-3

CMPT 880-1 G100" Special Topics in Medical Image Analysis
Instructor: G. Hamarneh SFU Burnaby

OBJECTIVE/DESCRIPTION:

Anatomical and functional medical imaging modalities (such as MRI, X-ray
CT, Ultrasound, PET, SPECT, and fMRI) are allowing us to peer inside the
human body and providing scientists and physicians with a wealth of
information indispensable for understanding, modelling, diagnosis and
treatment of diseases. The number of medical images acquired on a daily
basis is growing, finer details are captured and larger image sizes are
produced, and the dimensionality of images is increasing from two
dimensional scalar images to dynamic 3D multi-valued fields. This is
resulting in image data that cannot be effectively and accurately processed
with traditional simple visual inspection techniques. The development of
computational tools for medical image analysis therefore has tremendous
value and is the focus of this course. By reading and discussing recent
research papers on medical image analysis we will explore topics in medical
imaging, image processing/filtering, image segmentation, image registration
and shape modeling in the context of different applications such as
computer aided diagnosis and statistical shape analysis.

TOPICS:

o Medical imaging
o Medical Image processing/filtering
o Medical image segmentation
o Medical image registration
o Deformable shape modeling

GRADING:

To be discussed in the first week of classes. Grading will be primarily
based on students active participation in the discussions, paper
presentations, homework assignments and a project.

Students must attain an overall passing grade on the weighted average of
exams in the course in order to obtain a clear pass (C or better).

TEXTBOOKS:

o None

PREREQUISITES/COREQUISITES:

Interest in medical image analysis. Courses in image processing, computer
vision, or computer graphics would be an asset, but are not required.
Please feel free to contact me if you have any concerns about your
suitability for taking this course.