US 11,941,814 B2
Auto segmentation using 2-D images taken during 3-D imaging spin
Neil R. Crawford, Chandler, AZ (US); and Norbert Johnson, North Andover, MA (US)
Assigned to Globus Medical Inc., Audubon, PA (US)
Filed by GLOBUS MEDICAL, INC., Audubon, PA (US)
Filed on Nov. 4, 2020, as Appl. No. 17/088,975.
Prior Publication US 2022/0138952 A1, May 5, 2022
Int. Cl. G06K 9/00 (2022.01); A61B 6/03 (2006.01); G06T 7/11 (2017.01)
CPC G06T 7/11 (2017.01) [A61B 6/032 (2013.01); A61B 6/4028 (2013.01); A61B 6/4085 (2013.01); A61B 6/505 (2013.01); G06T 2207/10081 (2013.01); G06T 2207/30012 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method of identifying and segmenting anatomical structures from cone beam CT images, the method comprising:
receiving, from a cone beam CT device, at least one x-ray image, which is part of a plurality of x-ray images taken from a 360 degree scan of a patient, the at least one x-ray image containing at least one anatomical structure;
identifying and segmenting the at least one anatomical structure contained in the x-ray image based on a stored model of anatomical structures; and
creating a 3-D image volume from a plurality of x-ray images from the 360 degree scan;
adding the identification and segmentation information derived from the at least one x-ray image to the created 3-D image volume,
the step of receiving includes receiving a set of x-ray images at regularly spaced angular orientations; and
for each received x-ray image,
determining a confidence level of a match for the each x-ray image;
determining optimal identification and segmentation information based on the confidence levels
wherein the step of determining optimal identification and segmentation information includes determining the center of the anatomical structure by weighting the x-ray images based on the confidence level.