OATAO - Open Archive Toulouse Archive Ouverte Open Access Week

3D+t segmentation of PET images using spectral clustering

Zbib, Hiba and Kdouh, Salam and Mouysset, Sandrine and Stute, Simon and Girault, Jean-Marc and Charara, Jamal and Nassereddme, Mohammad and Mcheik, Ali and Buvat, Irène and Tauber, Clovis 3D+t segmentation of PET images using spectral clustering. (2015) In: 3rd International Conference on Advances in Biomedical Engineering (ICABME 2015), 16 September 2015 - 18 September 2015 (Beirut, Lebanon).

[img]
Preview
(Document in English)

PDF (Author's version) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
492kB

Official URL: http://dx.doi.org/10.1109/ICABME.2015.7323248

Abstract

Segmentation of dynamic PET images is often needed to extract the time activity curve (TAC) of regions. While clustering methods have been proposed to segment the PET sequence, they are generally either sensitive to initial conditions or favor convex shaped clusters. Recently, we have proposed a deterministic and automatic spectral clustering method (AD-KSC) of PET images. It has the advantage of handling clusters with arbitrary shape in the space in which they are identified. While improved results were obtained with AD-KSC compared to other methods, its use for clinical applications is constrained to 2D+t PET data due to its computational complexity. In this paper, we propose an extension of AD-KSC to make it applicable to 3D+t PET data. First, a preprocessing step based on a recursive principle component analysis and a Global K-means approach is used to generate many small seed clusters. AD-KSC is then applied on the generated clusters to obtain the final partition of the data. We validated the method with GATE Monte Carlo simulations of Zubal head phantom. The proposed approach improved the region of interest (ROI) definition and outperformed the K-means algorithm.

Item Type:Conference or Workshop Item (Paper)
Additional Information:Thanks to IEEE editor. The definitive version is available at http://ieeexplore.ieee.org This papers appears in Proceedings of ICABME 2015. ISBN : 978-1-4673-6516-1 The original PDF of the article can be found at : http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=7323248&tag=1
HAL Id:hal-01282045
Audience (conference):International conference proceedings
Uncontrolled Keywords:
Institution:French research institutions > Commissariat à l'Energie Atomique et aux énergies alternatives - CEA (FRANCE)
French research institutions > Centre National de la Recherche Scientifique - CNRS (FRANCE)
Université de Toulouse > Institut National Polytechnique de Toulouse - INPT (FRANCE)
French research institutions > Institut National de la Santé et de la Recherche Médicale - INSERM (FRANCE)
Other partners > Lebanese University - LU (LEBANON)
Université de Toulouse > Université Toulouse III - Paul Sabatier - UPS (FRANCE)
Université de Toulouse > Université Toulouse - Jean Jaurès - UT2J (FRANCE)
Université de Toulouse > Université Toulouse 1 Capitole - UT1 (FRANCE)
Other partners > Centre Hospitalier Régional Universitaire de Tours - CHU Tours (FRANCE)
Other partners > Université François-Rabelais de Tours (FRANCE)
Other partners > Université Paris-Sud 11 (FRANCE)
Laboratory name:
Statistics:download
Deposited By: IRIT IRIT
Deposited On:09 Feb 2016 15:47

Repository Staff Only: item control page