BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Geodesic methods for Biomedical Image Segmentation
- Cohen\, L (Universit Paris-Dauphine)
DTSTART;TZID=Europe/London:20110825T090000
DTEND;TZID=Europe/London:20110825T094500
UID:TALK32492AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/32492
DESCRIPTION:Tubular and tree structures appear very commonly i
n biomedical images like vessels\, microtubules or
neuron cells. Minimal paths have been used for lo
ng as an interactive tool to segment these structu
res as cost minimizing curves. The user usually pr
ovides start and end points on the image and gets
the minimal path as output. These minimal paths co
rrespond to minimal geodesics according to some ad
apted metric. They are a way to find a (set of) cu
rve(s) globally minimizing the geodesic active con
tours energy. Finding a geodesic distance can be s
olved by the Eikonal equation using the fast and e
fficient Fast Marching method. In the past years w
e have introduced different extensions of these mi
nimal paths that improve either the interactive as
pects or the results. For example\, the metric can
take into account both scale and orientation of t
he path. This leads to solving an anisotropic mini
mal path in a 2D or 3D+radius space. On a differen
t level\, the user interaction can be minimized by
adding iteratively what we called the keypoints\,
for example to obtain a closed curve from a singl
e initial point. The result is then a set of minim
al paths between pairs of keypoints. This can also
be applied to branching structures in both 2D and
3D images. We also proposed different criteria to
obtain automatically a set of end points of a tre
e structure by giving only one starting point. Mor
e recently\, we introduced a new general idea that
we called Geodesic Voting or Geodesic Density. Th
e approach consists in computing geodesics between
a given source point and a set of points scattere
d in the image. The geodesic density is defined at
each pixel of the image as the number of geodesic
s that pass over this pixel. The target structure
corresponds to image points with a high geodesic d
ensity. We will illustrate different possible appl
ications of this approach. The work we will presen
t involved as well F. Benmansour\, Y. Rouchdy and
J. Mille at CEREMADE.\n
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:Mustapha Amrani
END:VEVENT
END:VCALENDAR