Kolloquium Sommersemester 2008
Die Dozenten der Informatik
Dr. Kazunori Okada, San Francisco State University
spricht über
Robust Medical Image Analysis using the Mean Shift Framework
| Datum: | Montag, 26. Mai 2008
| | Zeit: | 17:15 Uhr
| | Ort: | Hörsaal 1, Römerstraße 164
|
ZusammenfassungIn past decades, medical imaging devices have drastically improved their qualities by rapidly increasing image resolutions, aiming for better diagnostic accuracy. However, this also caused an exponential increase in the amount of data that radiologists must evaluate, causing ever-increasing labor burdens and higher chances for misdiagnoses due to fatigues. Medical image analysis has been a potential answer to such critical demands.
In this talk, I will present my work toward making such medical image analysis more robust, efficient, and interactive, in order to meet the above clinical and technical demands. Mean Shift is a robust feature space analysis framework which has been successfully applied to various computer vision problems. I will present extensions of the mean shift to Gaussian scale-space and maximum-a-posteriori estimation. These extended mean shift frameworks are then successfully applied to clinical applications for detecting and segmenting the lung cancers (nodules) in 3D CT scan data. The results of our experiments with clinically-significant and technically-challenging ground-glass-nodule (GGN) cases will be illustrated.
(Vortrag auf Einladung von D. Cremers)
Letzte Änderung:
5. Mai 2008, 12:00:37
|
.
Sommersemester 2008