Geometry planning and image registration in magnetic particle imaging using bimodal fiducial markers
| Autor: | René Werner, Kolja Them, Johannes Salamon, Franziska Werner, Tobias Knopp, Caroline Jung, Harald Ittrich, Michael G. Kaul, Dennis Säring, T Mummert, Oliver M. Weber, M Hofmann, Gerhard Adam |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Image fusion
Quantitative imaging medicine.diagnostic_test Computer science business.industry Radiography Image registration Computed tomography Magnetic resonance imaging Geometry General Medicine Imaging phantom Object detection 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Magnetic particle imaging 030220 oncology & carcinogenesis medicine Fiducial marker business Image resolution |
| Zdroj: | Medical Physics. 43:2884-2893 |
| ISSN: | 0094-2405 |
| DOI: | 10.1118/1.4948998 |
| Popis: | Purpose: Magnetic particle imaging (MPI) is a quantitative imaging modality that allows the distribution of superparamagnetic nanoparticles to be visualized. Compared to other imaging techniques like x-ray radiography, computed tomography (CT), and magnetic resonance imaging (MRI), MPI only provides a signal from the administered tracer, but no additional morphological information, which complicates geometry planning and the interpretation of MP images. The purpose of the authors’ study was to develop bimodal fiducial markers that can be visualized by MPI and MRI in order to create MP–MR fusion images. Methods: A certain arrangement of three bimodal fiducial markers was developed and used in a combined MRI/MPI phantom and also during in vivo experiments in order to investigate its suitability for geometry planning and image fusion. An algorithm for automated marker extraction in both MR and MP images and rigid registration was established. Results: The developed bimodal fiducial markers can be visualized by MRI and MPI and allow for geometry planning as well as automated registration and fusion of MR–MP images. Conclusions: To date, exact positioning of the object to be imaged within the field of view (FOV) and the assignment of reconstructed MPI signals to corresponding morphological regions has been difficult. The developed bimodal fiducial markers and the automated image registration algorithm help to overcome these difficulties. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text