Parent Logos CBL LOGO
 
 
 
 
 
 
 
 
 
 
COSC 6391
 

COSC 6391
Biomedical Image Analysis

http://www.cbl.uh.edu/~6391s08

                            T, Th: 1-2:30pm PGH 344
Department of Computer Science - Uiversity of Houston

Spring 2008

OVERVIEW

Have you ever wondered what role images can play in Medicine and Biology? What can a computer scientist do with images? Would you like a sneak peek into the future of Preemptive-Predictive-Personalized Medicine?

This course will focus on computational methods and techniques for processing and analysis of biomedical images. The mathematical concepts will be covered with emphasis on algorithms that can be used to analyze an image in terms of content and properties (e.g., what are their characteristics - size, shape, texture, color, motion) of objects in the image and provide syntactical interpretations. The link between theoretical fundamentals, algorithms, and real-life applications will be accomplished through hands-on team projects. Project emphasis will be around collaborative learning and knowledge exchange leading to innovative solutions to real-life problems in current medical practice.

NOTE: The Biomedical Image Analysis course has recently been approved as one of the courses that can count towards fulfilling the core requirements for PhD students.

TOPICS

  • Mathematical transforms for Biomedical image processing: Radon, Wavelets
  • Biomedical image transforms: image registration, fusion, similarity metrics, interpolation, and optimization
  • Biomedical image segmentation: variational, probabilistic, model-based, region-based, and integration
  • Biomedical image modeling and estimation: statistics- and physics-based models, probabilistic models, shape representation, motion estimation
  • Biomedical image visualization: 2D, 3D, 4D, and 5D interpretations, multispectral visualization, dimensionality reduction, projections
  • Validation and Evaluation in Biomedical Image Analysis

PREREQUISITES

    REQUIRED
  • Programming knowledge and experience in C++ and/or Java
  • Digital Image Processing or Biomedical Image Analysis
    DESIRABLE
  • Software Engineering knowledge
  • Pattern Recognition/Data Mining knowledge and experience
  • Numerical methods knowledge

TENTATIVE EVALUATION

Homeworks 30%
Class Participation & Presentation 10%
Group Project 60%

REFERENCE BOOKS

  • Image Processing, Analysis, and Machine Vision 2nd Edition by Milan Sonka
  • Handbook of Image and Video Processing, 2nd Edition by A.C. Bovik
  • Numerical Methods in Biomedical Engineering, by S.M. Dunn, A. Constantinides, and P.V. Moghe
  • Biomedical Image Analysis by R.M. Rangayyan

INSRUCTOR

Prof. Ioannis A. Kakadiaris
Department of Computer Science, University of Houston
209 Philip G. Hoffman (PGH) Hall
Tel: (713) 743-1255
Email: ioannisk@uh.edu
URL: http://www.cbl.uh.edu/~ioannisk

TA

Deepak Roy Chittajallu
Department of Computer Science, University of Houston
235 Philip G. Hoffman (PGH) Hall
Email: drchittajallu@uh.edu