gms | German Medical Science

55. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie e. V. (DGNC)
1. Joint Meeting mit der Ungarischen Gesellschaft für Neurochirurgie

Deutsche Gesellschaft für Neurochirurgie (DGNC) e. V.

25. bis 28.04.2004, Köln

Multisegmental image-fusion of the spine

Mehrsegmentale Bildfusion der Wirbelsäule

Meeting Abstract

  • corresponding author Jan Kaminsky - Abteilung für Neurochirurgie, Medizinische Hochschule Hannover, Hannover
  • T. Rodt - Abteilung für Neurochirurgie, Medizinische Hochschule Hannover, Hannover
  • F. Donnerstag - Abteilung für Neuroradiologie, Medizinische Hochschule Hannover, Hannover
  • J. Zajaczek - Abteilung für Neuroradiologie, Medizinische Hochschule Hannover, Hannover
  • M. Zumkeller - Abteilung für Neurochirurgie, Medizinische Hochschule Hannover, Hannover

Deutsche Gesellschaft für Neurochirurgie. Ungarische Gesellschaft für Neurochirurgie. 55. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie e.V. (DGNC), 1. Joint Meeting mit der Ungarischen Gesellschaft für Neurochirurgie. Köln, 25.-28.04.2004. Düsseldorf, Köln: German Medical Science; 2004. DocMI.04.01

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter: http://www.egms.de/de/meetings/dgnc2004/04dgnc0270.shtml

Veröffentlicht: 23. April 2004

© 2004 Kaminsky et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.de). Er darf vervielf&aauml;ltigt, verbreitet und &oauml;ffentlich zug&aauml;nglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.


Gliederung

Text

Objective

Fusion of medical images represents a technique that allows correlation of homologous anatomical structures in different imaging modalities by a spatial transformation of the data sets. CT and MR of the spine provide complimentary information that is relevant for diagnostic and therapeutic decisions. To facilitate the use and enhance the precision of CT-MR-fusion of the spine a multisegmental fusion algorithm was developed, taking the segmental structure of the spine into account.

Methods

Routine clinical CT and MR of different sections of the spine was obtained and transferred to a PC-workstation. Following segmentation of the CT-data using manual and thresholding techniques, landmarks for the individual motion segments were defined in the CT and MR data. The developed algorithm then performed a rigid registration of the CT information to the MR data by calculating rotation and translation for each individual motion segment. The fused datasets could be evaluated as colour-coded images or using dynamic variation of the transparency. To assess the registration precision fiducial registration error (FRE), fiducial localisation error (FLE) and target registration error (TRE) were calculated.

Results

The developed algorithm allowed robust multi-segmental image fusion of the spine. The average time effort for defining the landmarks was 22 s/landmark and 34 s/landmark for CT and MR respectively. Calculation of the transformation matrix took less than one second. The average FRE and FLE were 1.53 mm and 1.74 mm respectively. Colour-coded images had advantages when the contours of the anatomical structures had to be assessed, whereas dynamic variation of the transparency allowed a better overall assessment of the spatial relationship of the anatomical structures.

Conclusions

The developed algorithm allows precise multi-segmental fusion of CT and MR of the spine. This could not be accomplished using current fusion-algorithms due to possible variation in the spatial orientation of the anatomical structures caused by motion artefacts, differing positioning of the patient and MR imaging artefacts. This technique could be beneficial for radiological diagnosis, surgical planning and navigated surgery.