Visualization with anatomical intelligence
A framework for facilitating visualization, including localizing at least one anatomical structure of interest in image data. The structure of interest is then highlighted by reformatting the image data by mapping landmarks associated with the structure of interest to corresponding points along a contour of a geometric shape and warping the image data based on the mapped landmarks. The resulting reformatted image data is rendered for display to a user.
1. A computer-implemented method of visualization, comprising:
(i) receiving magnetic resonance image data;
(ii) automatically localizing at least one bone structure of interest in the image data that appears only in a first number of slices in the image data;
(iii) highlighting and flattening the localized bone structure of interest by reformatting the image data into reformatted image data by
estimating a cylindrical shape of the localized bone structure of interest based on landmarks associated with the localized bone structure of interest,
resampling voxels on a surface of the cylindrical shape from the image data, and
mapping the resampled voxels on the cylindrical shape to corresponding points on a visualization plane to flatten the localized bone structure of interest, and
wherein the bone structure of interest appears only in a second number of slices of the reformatted image data, wherein the second number of slices is less than the first number of slices; and
(iv) rendering the reformatted image data for display via a display device.
2. The method of claim 1 wherein automatically localizing the bone structure of interest comprises performing a segmentation technique and generating a segmentation mask that delineates the bone structure of interest.
3. The method of claim 2 wherein the segmentation technique comprises atlas-based segmentation, deformable model-based segmentation, classification-based tissue labeling, or a combination thereof.
4. The method of claim 2 wherein highlighting the localized bone structure of interest comprises applying the segmentation mask to the image data and removing structures outside the mask.
5. The method of claim 1 wherein automatically localizing the bone structure of interest comprises detecting the landmarks associated with the localized bone structure of interest.
6. The method of claim 5 wherein detecting the landmarks associated with the localized bone structure of interest comprises performing learning-based detection, salient point detection, or a combination thereof.
7. The method of claim 1 wherein highlighting and flattening the localized bone structure of interest comprises estimating multiple cylindrical shapes of the bone structure of interest based on the landmarks associated with the localized bone structure of interest and mapping voxels on the multiple cylindrical shapes to corresponding visualization planes.
8. The method of claim 1 wherein the cylindrical shape comprises an ellipse cylinder or a circle cylinder.
9. The method of claim 1 wherein rendering the reformatted image data for display via the display device comprises performing multi-planar reformatting (MPR), maximum/minimum intensity projection (MIP) and Volume Rendering Technique (VRT) based on the reformatted image data.
10. A non-transitory computer readable medium embodying a program of instructions executable by machine to perform steps for visualization, the steps comprising:
(i) receiving magnetic resonance image data;
(ii) automatically localizing at least one bone structure of interest in the image data that appears only in a first number of slices in the image data;
(iii) highlighting and flattening the localized bone structure of interest by reformatting the image data into reformatted image data by
estimating a cylindrical shape of the localized bone structure of interest based on landmarks associated with the localized bone structure of interest,
resampling voxels on a surface of the cylindrical shape from the image data,
mapping the resampled voxels on the cylindrical shape to corresponding points on a visualization plane to flatten the localized bone structure of interest, and
wherein the bone structure of interest appears only in a second number of slices of the reformatted image data, wherein the second number of slices is less than the first number of slices; and
(iv) rendering the reformatted image data for display via a display device.