<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
<div class="moz-text-html" lang="x-western"> <tt>Yoav Freund will be
giving a talk this Friday, July 14 at 2:30pm in
room 402 of the CS building. Title and abstract follow below.<br>
<br>
</tt>
<hr size="2" width="100%"><tt><br>
Title: Using adaptive computer vision to analyze biomedical images<br>
Speaker: Yoav Freund<br>
Computer Science and Engineering, University of California,
San Diego<br>
<a class="moz-txt-link-freetext"
href="http://www.cse.ucsd.edu/%7Eyfreund">http://www.cse.ucsd.edu/~yfreund</a><br>
Date: Friday, July 14, 2:30pm<br>
Place: Room 402, CS building<br>
Host: Rob Schapire<br>
<br>
Abstract:<br>
</tt>
<blockquote><tt>The rapid advancement of biomedical imaging techniques
such as fluorescence microscopy, non-linear optics, MRI and ultra-sound
make it now possible to observe the time evolution of mechanical and
biochemical processes in cells and in the human body at a resolution </tt><tt>unimaginable
a decade ago. This ability to measure opens up an incredible range of
new possibilities for research in biology and in medicine. However, a
major challenge along the way is the formidable amount of manual work
required to extract meaningful quantitative results from the enormous
amounts of collected image data.</tt><br>
<br>
<tt>I suggest a combination of computer vision and machine learning
techniques to tackle such problems. In this talk I describe some of my
current work applying these ideas in several projects:</tt><br>
<blockquote>
<ol>
<li><tt>The study of the dynamics of lamellipodia in spreading
cells.</tt></li>
<li><tt>The study of regulatory networks in the development of
drosophila embryos.</tt></li>
<li><tt>Detection of protein micro-crystals for X-ray
crystalography.</tt></li>
<li><tt>Detection and grading of prostate cancer using microscope
images of biopsies.</tt><br>
</li>
</ol>
</blockquote>
<tt>I will conclude with some of the lessons I have learned from
these projects and some words on my current approach to automating
image analysis tasks.<br>
</tt></blockquote>
<br>
<hr size="2" width="100%"></div>
<br>
<pre class="moz-signature" cols="72">--
Princeton University
Department of Computer Science
35 Olden Street
Princeton, NJ 08540 USA
tel: +1 609 258 7726 fax: +1 609 258 1771
Computer Science Building, Room 407
<a class="moz-txt-link-freetext" href="http://www.cs.princeton.edu/~schapire">http://www.cs.princeton.edu/~schapire</a>
<a class="moz-txt-link-abbreviated" href="mailto:schapire@cs.princeton.edu">schapire@cs.princeton.edu</a>
</pre>
</body>
</html>