Three-dimensional reconstruction of a vascular network by dynamic tracking of magnetite nanoparticles

Authors

  • Olamaei Nina,

    1. NanoRobotics Laboratory, Ecole Polytechnique Montréal, 2500 Chemin de Polytechnique, Montreal, Quebec H3T 1J4, Canada and Imaging and 4D Visualization Laboratory, Ecole Polytechnique Montréal, 2500 Chemin de Polytechnique, Montreal, Quebec H3T 1J4, Canada
    Search for more papers by this author
  • Cheriet Farida,

    1. Imaging and 4D Visualization Laboratory, Ecole Polytechnique Montréal, 2500 Chemin de Polytechnique, Montreal, Quebec H3T 1J4, Canada
    Search for more papers by this author
  • Deschênes Sylvain,

    1. Department of Medical Imaging, Centre Hospitalier Universitaire Sainte-Justine, 3175 Chemin de la Côte Sainte-Catherine, Montreal, Quebec H3T 1C5, Canada
    Search for more papers by this author
  • Sharafi Azadeh,

    1. NanoRobotics Laboratory, Ecole Polytechnique Montréal, 2500 Chemin de Polytechnique, Montreal, Quebec H3T 1J4, Canada
    Search for more papers by this author
  • Martel Sylvain

    1. NanoRobotics Laboratory, Ecole Polytechnique Montréal, 2500 Chemin de Polytechnique, Montreal, Quebec H3T 1J4, Canada
    Search for more papers by this author

Abstract

Purpose:

Visualization of small blood vessels feeding tumor sites provides important information on the tumors and their microenvironment. This information plays an important role in targeted drug therapies using magnetic gradients. However, capabilities of current clinical imaging modalities may be insufficient to resolve complex microvascular networks. The purpose of this study is to map the vascular network, 3D, based on the magnetic susceptibility contrast.

Methods:

Magnetic particles induce an inhomogeneity in the MRI's magnetic field in an order much larger than their real size. This is an approach to compensate the spatial resolution insufficiency of a clinical MR scanner. Micron-sized agglomerations of magnetite nanoparticles were injected in a 3D phantom vascular network, and a fast multislice, multiacquisition MR sequence was applied to track the agglomerations along their trajectories. The experiment was performed twice for two different imaging planes: coronal and transversal. The susceptibility artifact in the images indicated the presence and the position of the agglomerations. The calculated positions through multiple images were assembled to build up the 3D distribution of the vascular network.

Results:

The calculated points were compared with the centerline of the channels, extracted from the 3D reference image, to determine the absolute measurement error. The mean error was measured to be approximately half of the pixel's size. It was found that the positioning error on the axis perpendicular to the imaging slice was nearly twice as high as on the imaging plane axes due to the slice thickness. In order to compensate for the lack of resolution on the perpendicular axis, the reconstruction was performed using a combination of coronal and transversal data. The combination of the coordinates led to a significant decrease in the mean measurement error at each segment in the vascular network (p < 0.001).

Conclusions:

A method for 3D reconstruction of a microvascular network based on the susceptibility contrast in MRI and using a clinical scanner and a commercial receiver coil was proposed. The method presents a novel approach for reconstruction of vascular networks using the susceptibility effect. The proposed method may be applied to resolve vascular networks at a micrometric scale.

Ancillary