Für eine korrekte Darstellung dieser Seite benötigen Sie einen XHTML-standardkonformen Browser, der die Darstellung von CSS-Dateien zulässt.


Automated Image Analysis in Computer-Aided Multiple Sclerosis Diagnosis

Dr. Jacek Kawa, Silesian University of Technology

Date: 15/05/2013
Time: 10:15-11:15
Room: H-C 3303

Multiple Sclerosis (MS) is a chronic demyelinating disease of a brain and spinal cord and is a major cause of disability. The progress of the disease may be slow down or stopped if a correct treatment is started – hence the reliable diagnostic and follow-up monitoring is important and a medical imaging (Magnetic Resonance Imaging, MRI) is used. In MRI multiple, confluent demyelination lesions are visible mostly within white matter of brain.
Quantitative assessment of a demyelination process is infeasible without computer image analysis tools. The lesions are located in various regions of the brain and are relatively small. Thus a lot of algorithms has been developed worldwide to help in segmentation and monitoring.
During the lecture several approaches will be presented to the MRI computer aided diagnostic and follow-up. Uses of T1-wi, T2-wi, PD and FLAIR MRI images will be discussed. Various processing pipelines will be introduced. Simple semi-automatic region-growing methods will be introduced first. Multi-dimensional histogram and coincidence matrix algorithms will be presented later. Finally, fully automated, kernelized clustering and fuzzy connectedness-based methods will be shown. The output parameters will be discussed from a diagnostic point of view.
Evaluation of results will be shown.