Systems Bioimaging Lab

ECE594Q: Computational Bio-Microscopy (Winter 2008)

Home / Teaching / ECE594Q

Home
News
People
Research
Publications
Software/Demos
Teaching
Syllabus
Grading
Office Hours
Links
<--> Resources

ECE 594Q Computational Bio-Microscopy, Winter 2008

Prof. Michael Liebling
Department of Electrical on Computer Engineering
University of California Santa Barbara

Hours
Tuesday 10–11:50am, Phelps Hall 1437
Thursday 10–11:50am, Phelps Hall 1437

Office Hours 
Harold Frank Hall 3155
Tue 4-5 pm
Thu 4-5 pm (Updated)
or by appointment: Michael Liebling, liebling AT ece.ucsb.edu
(please put ECE594Q in the subject line)

Teaching Assistants
TBD

Introduction
Microscopy has become a central tool for analysis and diagnosis in biology and medicine. Modern microscopy systems increasingly rely on sophisticated digital signal processing techniques for image formation, enhancement, and analysis and require the synthesis of multiple disciplines that include biology, optics, and computational techniques.

Course Objectives

• Provide an introduction to the working principle of common and emerging bio-microscopy techniques from a signal processing perspective and compare features and limitations of optical imaging techniques
• Introduce commonly used signal processing algorithms for image formation and enhancement in microscopy.
• Prepare students for interdisciplinary collaborations by:
  • introducing them to methodologies underlying biological/medical studies and how they can be translated into concrete engineering problems
  • providing an overview of the most commonly used biological sample preparation techniques

Topics to include
Anatomy of a Microscope (widefield, confocal and multi-photon microscopy), aberrations, linear systems and Fourier optics, fluorescence (features, limitations, and applications), spectral unmixing, colocalization, deconvolution, digital holographic microscopy, superresolution (structured illumination, photo-activated localization microscopy, stimulated emission depletion microscopy), adaptive optics.

Form

• Lecture
• Homework: theory and Matlab programming exercises based on real-world datasets
• Journal paper reading and critique: 1 interdisciplinary biology, signal processing or optics techniques paper

Requirements
Signal processing, Fourier transform, FFT, convolution, linear systems. Linear algebra. Interest in mathematics, optics, and biology.

Schedule (Subject to Change)  

Recommended Textbooks
E. Meijering and G. van Cappellen, ``Biological Image Analysis Primer,'' Erasmus MC, Rotterdam, 2006, http://www.imagescience.org/meijering/\smallskip\\<--> M. Müller, "Introduction to Confocal Fluorescence Microscopy," 2nd Edition, SPIE Press (Bellingham, WA), 2005 (SPIE members get the book for $39) ISBN 0-8194-6043-5
J. W. Goodman" Introduction to Fourier Optics" 3rd Edition, Roberts and Company Publishers (Greenwood Village, CO), 2004. ISBN: 978-0974707723

See also the books on the Course Reserve at the UCSB library: ECEN594Q-LIB Bibliography: Goodman Bertero Fractional Fourier Transform Introduction to confocal fluorescence microscopy (Muller) -->

Homework
HW1-3 will each have two short problems related to the course. HW4-6 will each consist in writing a short Matlab (typically 10-20 lines) program and apply it to experimental or simulated data sets. On their due date, homeworks must be handed in at the beginning of class. Since the solutions will be discussed in class, no late homeworks will be accepted. Homeworks that were not turned in will be assigned a 0 point count. Only the five highest homework grades will be taken into account to compute the average homework grade, therefore, there will be no make-up assignments (see Grading, below). Early submission are accepted if student cannot be present at beginning of class (must be arranged with instructor).

Midterm Exam
The midterm will consist of a problem set, similar to HW1-3 and some general questions about concepts discussed in the class. The midterm is a closed-book exam.

Final
Each student chooses a technique paper in (broad) relation to the topics covered in class. Paper must be either in list of proposed papers or approved by instructor. Students are encouraged to choose and read their paper as early as possible. They must indicate their choice (by email) to the instructor on Tuesday January 29 at the latest, as well as short answers to (or how they intend to approach) the following questions in their final paper:

• Context (What problem does this technique address?)
• Method description (What is the proposed solution to the problem?)
• Features and limitations (What are the pitfalls of the proposed method? Have any improvements been proposed subsequently? What are your ideas on improving the method?)
• Applications (Search pubmed/ISI for papers that apply this method in a biological context or give suggestions of possible use in biology).
  
  Please refer to the detailed instructions and the list of papers.

Grading 
All work will be graded on a point scale from 0-100.
The final course grade will be computed as follows:
    Total Point = 50% Homework + 20% Midterm exam + 30% Final Paper
where the lowest homework grade does not contribute to the Homework average, that is:
    Homework = ∑_i∈S h_i/(N-1),
where S = {1,...,N} \ i_joker, with i_joker = arg min_i h_i, and h_i is the grade for homework i=1,...,N, N=6. Total Point will then be converted to a letter grade or satisfactory/unsatisfactory:     

Links 
Workshop on Bio-Image Informatics 2008 @ UCSB
Microscopy Primer
Microscopy from the Beginning (Zeiss)
ImageJ
Imaris (Bitplane AG)

Latest Updates