Conformal voxel magnetic resonance spectroscopy
In the past five years, there has been an explosion in the number of clinical institutions using in vivo magnetic resonance spectroscopy (MRS) for the identification and grading of brain tumours, white matter disease, degenerative disease, prostate cancer, and many other pathologies. Working off a magnetic resonance image, spectroscopists can now select a cube-shaped area (called a voxel) inside the tissue of interest and acquire an MR spectrum from just that area.
The method, however, is not without shortcomings. Tumours, for example, do not grow in the shape of a cube. Therefore when defining the area to examine, the spectroscopist must make the choice either to exclude pieces of a tumour by placing the voxel inside the tumour, or include normal tissue outside of the tumor by defining a voxel that encompasses the tumour. Neither choice is ideal as each risks measuring a spectrum that does not represent of the tumour.
NRC researchers have developed a solution to this problem by changing the way a voxel is prescribed. The solution, called conformal voxel MRS (CV-MRS) is not limited to cuboidal voxels. Instead, it uses a computer algorithm to optimally sculpt a voxel using many sharp spatial saturation slices to conform to the natural shape of the tumour, prostate, or other tissue of interest. In its simplest implementation, a technologist would simply click on the tissue of interest and the computer would do the rest. In addition to optimizing the spectrum obtained, CV-MRS essentially removes operator-to-operator differences by removing subjective decisions.
This technology is available for licensing, or for further development through a collaborative research agreement with NRC. The business opportunity may be referred to by its NRC ID: 11449.
- Automatic voxel definition maximizes the volume of tissue of interest while minimizing the volume of exclusion.
- Non-subjective voxel definition removes operator-to-operator differences.
- CV-MRS is user friendly.
The technology is available for exclusive or non-exclusive license.
NRC file 11449: Patent issued in the United States.
Claude Larose, Portfolio Business Advisor
- Date modified: