Biomedical
Imaging
Model-based Segmentation of Biomedical Images
Data Analysis in Small Animal and Clinical Medical Image Data
The
ISMV group has developed model-based 2D and 3D segmentation software
to enable biologists to automatically screen mouse image data to
find phenotypes of interest. This work has been accomplished in
collaboration with Texas Tech University with funding from ORNL,
the National Institutes of health and the Department of Energy. These advanced segmentation tools were primarily developed for
small animal imaging data analysis such as rapid phenotype screening
and anatomic-based quantification of isotope from nuclear images
(e.g. single photon emission CT or SPECT). The images on the right
show 2D and 3D results from model based segmentation tools applied
to small animal micro-CT data sets.
ISMV
is also applying these same image segmentation tools to clinical
medical image analysis problems. We are working with Columbia
University and Texas Tech University to extend our image processing
tools to assist doctors in mapping out the various functional
regions of the brain in images produced by magnetic resonance imaging
(MRI) systems. The brain MRI image data is overlaid with nuclear
images generated by a positron emission tomography (PET) system
to help the doctors quantify the effectiveness of new drug delivery
systems for the brain. Preliminary results of the 2D and 3D algorithms
as applied to human brain MRI images are shown in the bottom figure.
  |
| MRI
human brain segmentation: striatum in 2D (left) and cerebellum
in 3D (right). |
Capabilities
and Tools
- 2D and 3D Model-based Data Segmentation
- Phenotype Screening
- Computer-Aided Diagnosis
Fact
Sheet available here in PDF format.
Point
of Contact:
Jeffery R. Price, Ph.D.
R&D Staff, Image Science & Machine Vision
Engineering Science and Technology Division
Oak Ridge National Laboratory
P.O. Box 2008, MS-6010
Oak Ridge, Tennessee 37831-6010
Office: (865) 574-5743
E-mail: pricejr@ornl.gov |
