2
|
Layer T, Blaickner M, Knäusl B, Georg D, Neuwirth J, Baum RP, Schuchardt C, Wiessalla S, Matz G. PET image segmentation using a Gaussian mixture model and Markov random fields. EJNMMI Phys 2015; 2:9. [PMID: 26501811 PMCID: PMC4545759 DOI: 10.1186/s40658-015-0110-7] [Citation(s) in RCA: 10] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/25/2014] [Accepted: 09/08/2014] [Indexed: 12/05/2022] Open
Abstract
Background Classification algorithms for positron emission tomography (PET) images support computational treatment planning in radiotherapy. Common clinical practice is based on manual delineation and fixed or iterative threshold methods, the latter of which requires regression curves dependent on many parameters. Methods An improved statistical approach using a Gaussian mixture model (GMM) is proposed to obtain initial estimates of a target volume, followed by a correction step based on a Markov random field (MRF) and a Gibbs distribution to account for dependencies among neighboring voxels. In order to evaluate the proposed algorithm, phantom measurements of spherical and non-spherical objects with the smallest diameter being 8 mm were performed at signal-to-background ratios (SBRs) between 2.06 and 9.39. Additionally 68Ga-PET data from patients with lesions in the liver and lymph nodes were evaluated. Results The proposed algorithm produces stable results for different reconstruction algorithms and different lesion shapes. Furthermore, it outperforms all threshold methods regarding detection rate, determines the spheres’ volumes more accurately than fixed threshold methods, and produces similar values as iterative thresholding. In a comparison with other statistical approaches, the algorithm performs equally well for larger volumes and even shows improvements for small volumes and SBRs. The comparison with experts’ manual delineations on the clinical data shows the same qualitative behavior as for the phantom measurements. Conclusions In conclusion, a generic probabilistic approach that does not require data measured beforehand is presented whose performance, robustness, and swiftness make it a feasible choice for PET segmentation.
Collapse
Affiliation(s)
- Thomas Layer
- Institute of Telecommunications, Vienna University of Technology, Karlsplatz 13, Vienna, 1040 Wien, Austria. .,Health & Environment Department, Austrian Institute of Technology, Donau-City-Strasse 1/2, Vienna, 1220 Wien, Austria.
| | - Matthias Blaickner
- Health & Environment Department, Austrian Institute of Technology, Donau-City-Strasse 1/2, Vienna, 1220 Wien, Austria.
| | - Barbara Knäusl
- Department of Radiation Oncology, Division of Medical Radiation Physics, Christian Doppler Laboratory for Medical Radiation Research for Radiation Oncology, Medical University of Vienna/AKH Vienna, Währinger Gürtel 18-20, Vienna, 1090 Wien, Austria.
| | - Dietmar Georg
- Department of Radiation Oncology, Division of Medical Radiation Physics, Christian Doppler Laboratory for Medical Radiation Research for Radiation Oncology, Medical University of Vienna/AKH Vienna, Währinger Gürtel 18-20, Vienna, 1090 Wien, Austria.
| | - Johannes Neuwirth
- Radiation Safety and Applications, Seibersdorf Labor GmbH, 2444 Seibersdorf, Seibersdorf, Austria.
| | - Richard P Baum
- THERANOSTICS Center for Molecular Radiotherapy and Molecular Imaging (PET/CT) ENETS Center of Excellence, Zentralklinik Bad Berka, Robert-Koch-Allee 9, 99437 Bad Berka, Bad Berka, Germany.
| | - Christiane Schuchardt
- THERANOSTICS Center for Molecular Radiotherapy and Molecular Imaging (PET/CT) ENETS Center of Excellence, Zentralklinik Bad Berka, Robert-Koch-Allee 9, 99437 Bad Berka, Bad Berka, Germany.
| | - Stefan Wiessalla
- THERANOSTICS Center for Molecular Radiotherapy and Molecular Imaging (PET/CT) ENETS Center of Excellence, Zentralklinik Bad Berka, Robert-Koch-Allee 9, 99437 Bad Berka, Bad Berka, Germany.
| | - Gerald Matz
- Institute of Telecommunications, Vienna University of Technology, Karlsplatz 13, Vienna, 1040 Wien, Austria.
| |
Collapse
|