Towards iron oxide nanoparticles quantization in molecular MR images by default field deconvolution

  • Authors:

  • Delphine Charpigny;Thomas Grenier;Christophe Odet;Hugues Benoit-Cattin

  • Affiliations:

  • CREATIS-LRMN, INSA de lyon, Université de Lyon, CNRS UMR, Villeurbanne, France;CREATIS-LRMN, INSA de lyon, Université de Lyon, CNRS UMR, Villeurbanne, France;CREATIS-LRMN, INSA de lyon, Université de Lyon, CNRS UMR, Villeurbanne, France;CREATIS-LRMN, INSA de lyon, Université de Lyon, CNRS UMR, Villeurbanne, France

  • Venue:
  • ISBI'09 Proceedings of the Sixth IEEE international conference on Symposium on Biomedical Imaging: From Nano to Macro
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

MRI contrast agents, such as USPIO, are often used for pathology diagnoses. Locating and quantifying USPIO is of high interest. Contrast techniques like T2* cartography and positive contrast imaging have been developped in recent years. Another strategy is to analyse magnetic field inhomogeneities produced by the particles. We present a 3D post-processing method SYMDEF. This method consists in obtaining a SusceptibilitY Map by the deconvolution of DEfault Field map (the magnetic susceptibility is linked to USPIO concentration). Constrained least squares filtering is proposed to add robustness to SYMDEF. SYMDEF is compared with another method detailed by Haacke. Results show that SYMDEF allows us to retrieve susceptibility information from default field map without any prior knowledge as susceptibility. Our approach is a good candidate for USPIO quantization.