US 11,995,820 B2
Method and apparatus for automated detection of landmarks from 3D medical image data based on deep learning
Youngjun Kim, Seoul (KR); and Hannah Kim, Seoul (KR)
Assigned to IMAGOWORKS INC., Seoul (KR)
Appl. No. 17/295,916
Filed by IMAGOWORKS INC., Seoul (KR)
PCT Filed May 26, 2020, PCT No. PCT/KR2020/006789
§ 371(c)(1), (2) Date May 21, 2021,
PCT Pub. No. WO2021/210723, PCT Pub. Date Oct. 21, 2021.
Claims priority of application No. 10-2020-0045924 (KR), filed on Apr. 16, 2020.
Prior Publication US 2023/0024671 A1, Jan. 26, 2023
Int. Cl. G06T 7/00 (2017.01); G06T 15/08 (2011.01); G06T 15/20 (2011.01); G06V 10/25 (2022.01); G06V 10/82 (2022.01)
CPC G06T 7/0012 (2013.01) [G06T 15/08 (2013.01); G06T 15/20 (2013.01); G06V 10/25 (2022.01); G06V 10/82 (2022.01); G06T 2207/30008 (2013.01); G06T 2207/30036 (2013.01); G06V 2201/07 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A method for automated detection of landmarks from 3D medical image data using deep learning, the method comprises:
receiving a 3D volume medical image;
generating a 2D intensity value projection image based on the 3D volume medical image;
automatically detecting an initial anatomical landmark using a first convolutional neural network based on the 2D intensity value projection image;
generating a 3D volume area of interest based on the initial anatomical landmark; and
automatically detecting a detailed anatomical landmark using a second convolutional neural network different from the first convolutional neural network based on the 3D volume area of interest,
wherein an input data of the first convolutional neural network is the 2D intensity value projection image and an output data of the first convolutional neural network is the initial anatomical landmark, and
wherein an input data of the second convolutional neural network is the 3D volume area of interest and an output data of the second convolutional neural network is the detailed anatomical landmark.