Faculty of Informatics Vienna University of Technology Institute of Computer Aided Automation PRIP Home PRIP Home
Personal tools
You are here: Home Teaching Informatikpraktika Abschlusspräsentation Praktikum aus Computergraphik und digitaler Bildverarbeitung

Abschlusspräsentation Praktikum aus Computergraphik und digitaler Bildverarbeitung

*Automatic classification of the dignity of soft-tissue lesions using “Dynamic Contrast Enhanced Magnetic Resonance Imaging” (DCE-MRI)*

What
  • Praktika Presentation
When Jun 15, 2011
from 04:00 pm to 04:00 pm
Where Seminarraum 183/2
Add event to calendar vCal
iCal
*Abstract:*

*Automatic classification of the dignity of soft-tissue lesions using
“Dynamic Contrast Enhanced Magnetic Resonance Imaging” (DCE-MRI)*


This project is about clinical diagnostics of soft-tissue lesions using
medical imaging methods. Soft-tissue lesions are very uncommon and the
majority of these lesions are treated in Austria at the AKH Vienna. A
research project was started at the “Universitätsklinik für
Radiodiagnostik”, which analyses this kind of lesions in detail.

Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) is an imaging
method, which nowadays is commonly used for detecting lesions and assessing
their dignity. A contrast agent is injected in the area of investigation,
while MR images are acquired of the area in discrete time steps.

The contrast agent accumulates in the blood vessels. The blood supply to
malignant lesions is in common higher than the supply to the surrounding
tissue. So the contrast agent is accumulated in the lesion. This causes a
change in the intensity of the MR image. From the change of the intensity
over time a „time versus signal intensity“ curve can be created, which is
used for detecting lesions and classifying between benign and malign
lesions.

The aim of this project was to implement and train an automatic classifier,
which classifies soft-tissue lesions between benign and malign using the
„time versus signal intensity“ curve in the region of the lesion. A
sensitivity of over 90% and a specificity of around 45% was achieved.
Document Actions