Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Preoperative planning of neurovascular compression syndrome surgery based on an interactive wizard-based segmentation approach - extended evaluation results

 
: Franz, Daniela; Syré, Liane; Bischoff, Barbara; Hastreiter, Peter; Wittenberg, Thomas

Hahn, H.K. ; Deutsche Gesellschaft für Computer- und Roboterassistierte Chirurgie -CURAC-:
14. Jahrestagung der Deutschen Gesellschaft für Computer- und Roboterassistierte Chirurgie, CURAC 2015. Tagungsband : 17.-19.9.2015, Bremen
Bremen, 2015
ISBN: 978-3-00-050359-7
S.195-200
Deutsche Gesellschaft für Computer- und Roboterassistierte Chirurgie (CURAC Jahrestagung) <14, 2015, Bremen>
Englisch
Konferenzbeitrag
Fraunhofer IIS ()
Segmentierung; modellbasierte Segmentierung; medizinische BV; flächenbasierte Segmentierung; 3D Bildverarbeitung

Abstract
Compression of cranial nerves by a vessel at the entry or exit of the nerves to the brainstem may result in neurovascular compression syndrome (NVC). To enable a comprehensive spatial understanding of the patient specific anatomy, preoperative image-based planning of NVC surgery is mandatory. Distinctively, brainstem, vessels and nerves have to be delineated and visualized with adequate image analysis tools. In order to support this crucial image-processing task, we have recently proposed an interactive wizard-based segmentation method. This approach guides a physician through a set of successive segmentation steps, and for each step translates the related parameters into adequate medical terminology. The wizard also supports medical student to become acquainted with NVC related anatomy. After a pilot software validation study presented in previous work [1], we now have now applied the proposed interactive segmentation approach for NVCs to an extended set of 10 clinical cases and 9 users with clinical background. Especially, medical undergraduate students with no image analysis background showed a steep learning curve using the proposed wizard-based approach, and had needed in the mean only 30 minutes for the completion of the segmentation task for a NVC case.

: http://publica.fraunhofer.de/dokumente/N-377284.html