System and method for three-dimensional nerve segmentation using curved multiplanar reformatting magnetic resonance imaging
Systems and methods for segmenting a nerve in a three-dimensional image volume obtained with a magnetic resonance imaging (“MRI”) system are provided. A three-dimensional image volume that depicts a nerve and surrounding anatomical structures is provided and from that image volume the nerve is segmented. In general, a curved multiplanar reformatting (“CMPR”) process is utilized to mark, segment, and then display the nerve in three dimensions.
1. A method for segmenting a nerve in a three-dimensional image volume obtained with a magnetic resonance imaging (MRI) system, the steps of the method comprising:
(a) providing a three-dimensional image volume that depicts a nerve and surrounding anatomical structures in a series of contiguous image slices obtained with an MRI system;
(b) selecting a cross-sectional image from the image volume;
(c) generating a plurality of planes that intersect at a common point in the selected cross-sectional image and are orthogonal to the selected cross-sectional image;
(d) generating a planar reformatted image for each of the plurality of planes;
(e) estimating nerve contours in each of the planar reformatted images;
(f) receiving user feedback to adjust the estimated nerve contours;
(g) generating a surface estimate based on the adjusted nerve contours; and
(h) generating a volume that defines the nerve segmented from the surrounding anatomical structures based on the generated surface estimate.
2. The method as recited in claim 1 , wherein step (f) includes receiving user feedback to adjust a position of a point in the estimated nerve contour and recalculating the estimate of the nerve contour based on the received user feedback.
3. The method as recited in claim 1 , wherein step (f) includes adding at least one additional point to the estimated nerve contour based on the received user feedback.
4. The method as recited in claim 1 , wherein step (c) includes selecting a number of desired planes.
5. The method as recited in claim 1 , wherein step (g) includes generating an in-slice spline for each image slice based on an intersection of the estimated nerve contours with a given image slice.
6. The method as recited in claim 5 , wherein step (g) includes generating the surface estimate as a surface rendering based on the in-slice splines for each image slice.
7. The method as recited in claim 6 , wherein each surface rendering is generated based on intersection points defined by intersections of vectors extending from a center of the in-slice spline for a given image slice and the in-slice spline for that given image slice.
8. The method as recited in claim 7 , wherein each surface rendering is generated as a mesh triangulation based on the intersection points.