Cancer detection based on four quadrant mapping and matrix analysis of image data
A diagnostic system to analyze imaging data includes a memory configured to store hybrid imaging data of a tissue sample. The system also includes a processor operatively coupled to the memory and configured to generate a four quadrant plot based on the hybrid imaging data. Each point in the four quadrant plot corresponds to an image voxel of the tissue sample. The processor is also configured to determine one or more angle values and one or more distance values for image voxels in the four quadrant plot. The processor is further configured to identify one or more characteristics of the tissue sample based at least in part on the one or more angle values and the one or more distance values. The processor is further configured to perform a matrix analysis of the data, which can be used to identify the one or more characteristics of the tissue sample.
1 . A diagnostic system to analyze imaging data comprising:
a memory configured to store hybrid imaging data of a tissue sample; and
a processor operatively coupled to the memory and configured to:
generate a four quadrant plot based on the hybrid imaging data, wherein each point in the four quadrant plot corresponds to an image voxel of the tissue sample;
generate, based on the hybrid imaging data, a matrix of b-values and echo time (TE) values for each image voxel in the four quadrant plot;
determine, by the processor, a change in apparent diffusion coefficient (ADC) as a function of TE and a change in T 2 as a function of b-value based on the matrix;
identify, by the processor and based on the matrix, a plurality of eigenvalues for each image voxel; and
determine, by the processor and based at least in part on the plurality of eigenvalues, whether each image voxel of the tissue sample is associated with cancerous tissue or non-cancerous tissue.
2 . The system of claim 1 , further comprising a magnetic resonance imaging system configured to generate the hybrid imaging data, wherein the hybrid imaging data includes both T 2 -weighted imaging data and diffusion-weighted imaging data corresponding to the tissue sample.
3 . The system of claim 1 , wherein the matrix comprises a 3×3 matrix.
4 . The system of claim 1 , wherein a y-coordinate is equal to a slope of apparent diffusion coefficient (ADC) as a function of TE in the four quadrant plot.
5 . The system of claim 1 , wherein an x-coordinate is equal to a slope of T 2 as a function of b-value in the four quadrant plot.
6 . The system of claim 1 , wherein the processor determines one or more distance values that comprise absolute distances from an origin of the four quadrant plot to the image voxels.
7 . The system of claim 6 , wherein the processor determines one or more angle values for each of the image voxels, wherein the one or more angles values are relative to the origin of the four quadrant plot.
8 . The system of claim 1 , wherein the processor is further configured to generate one or more hybrid matrix analysis parameter plots based on the hybrid imaging data, wherein the one or more hybrid matrix parameter plots include the plurality of eigenvalues.
9 . The system of claim 8 , wherein the processor is configured to combine the plurality of eigenvalues and to map a combination of the plurality of eigenvalues to identify one or more characteristics of the tissue sample.
10 . The system of claim 1 , wherein determination of whether each image voxel of the tissue sample is associated with the cancerous tissue or the non-cancerous tissue is based at least in part on which quadrant the image voxels are positioned.
11 . A diagnostic method of analyzing imaging data, the method comprising:
storing, in a memory of a computing system, hybrid imaging data of a tissue sample; and
generating, by a processor operatively coupled to the memory, a four quadrant plot based on the hybrid imaging data, wherein each point in the four quadrant plot corresponds to an image voxel of the tissue sample;
generating, by the processor and based on the hybrid imaging data, a matrix of b-values and echo time (TE) values for each image voxel in the four quadrant plot;
determining, by the processor, a change in apparent diffusion coefficient (ADC) as a function of TE and a change in T 2 as a function of b-value based on the matrix;
identifying, by the processor and based on the matrix, a plurality of eigenvalues for each image voxel; and
determining, by the processor and based at least in part on the plurality of eigenvalues, whether each image voxel of the tissue sample is associated with cancerous tissue or non-cancerous tissue.
12 . The method of claim 11 , wherein y is equal to a slope of apparent diffusion coefficient (ADC) as a function of TE in the four quadrant plot, and x is equal to a slope of T 2 as a function of b-value in the four quadrant plot.
13 . The method of claim 11 , wherein the matrix comprises a 3×3 matrix.
14 . The method of claim 11 , further comprising combining and mapping, by the processor, the plurality of eigenvalues, wherein determination of whether each image voxel of the tissue sample is associated with the cancerous tissue or non-cancerous tissue is based at least in part on the mapped combination of eigenvalues.