US 11,836,926 B2
System and method for contouring a set of medical images based on deep learning algorithm and anatomical properties
Wonhoon Choi, Seoul (KR); Sang Wook Cho, Seoul (KR); and Han Young Kim, Seoul (KR)
Assigned to INFINITT HEALTHCARE CO., LTD., Seoul (KR)
Filed by INFINITT HEALTHCARE CO., LTD., Seoul (KR)
Filed on Jan. 20, 2021, as Appl. No. 17/153,174.
Claims priority of application No. 10-2020-0180306 (KR), filed on Dec. 21, 2020.
Prior Publication US 2022/0198674 A1, Jun. 23, 2022
Int. Cl. G06T 7/12 (2017.01); G06T 7/00 (2017.01)
CPC G06T 7/12 (2017.01) [G06T 7/0012 (2013.01); G06T 2200/24 (2013.01); G06T 2207/20104 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A system for contouring a set of medical images, the system comprising:
at least one processor;
a user interface; and
a communication interface,
wherein the at least one processor is configured to:
receive a medical image set including a plurality of medical images via the communication interface;
receive region-of-interest (ROI) information as an object for drawing of a contour via the user interface;
check whether a contour corresponding to the ROI information has been drawn on a target image currently displayed to a user;
predict a contour candidate in the target image if the contour corresponding to the ROI information has not been drawn on the target image; and
display the contour candidate on the target image,
wherein the contour candidate is predicted based on a relevance between a source image, on which a first contour corresponding to the ROI information has been already drawn, and the target image,
wherein the contour candidate is predicted such that a deformation of the contour candidate from the first contour has been constrained based on anatomical information related to the ROI information, and
wherein the deformation of the contour candidate from the first contour has been constrained based on a mechanical deformable property related to an anatomical classification of the ROI information.