US 12,329,581 B2
Method and system for generating a three-dimensional ultrasound image of a tissue volume from two-dimensional ultrasound images
Koosha Pourtahmasi Roshandeh, Alberta (CA); Dornoosh Zonoobi, Edmonton (CA); Abhilash Rakkunedeth, Alberta (CA); Masood Dehghan, Alberta (CA); and Jacob Jaremko, Alberta (CA)
Assigned to Exo Imaging, Inc.
Filed by Exo Imaging, Inc., Santa Clara, CA (US)
Filed on Sep. 1, 2023, as Appl. No. 18/460,158.
Application 18/460,158 is a continuation of application No. 17/294,689, granted, now 11,779,309, previously published as PCT/SG2019/050564, filed on Nov. 19, 2019.
Claims priority of application No. 10201810322Y (SG), filed on Nov. 19, 2018.
Prior Publication US 2024/0041432 A1, Feb. 8, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. A61B 8/00 (2006.01); G01S 15/89 (2006.01); G06N 3/02 (2006.01)
CPC A61B 8/483 (2013.01) [A61B 8/5207 (2013.01); G01S 15/8915 (2013.01); G01S 15/8993 (2013.01); G06N 3/02 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for generating a three-dimensional (3D) ultrasound image of a tissue volume using at least one processor, the method comprising:
generating a series of two-dimensional (2D) ultrasound images of the tissue volume associated with a plurality of positions, respectively, along a scanning direction of the tissue volume;
for a respective pair of consecutive 2D ultrasound images of the series of 2D ultrasound images:
generating a difference image from the respective pair of consecutive 2D ultrasound images;
estimating, using a deep neural network, an estimated distance between the positions associated with the respective pair of consecutive 2D ultrasound images based on a classification of the difference image, wherein the deep neural network is trained to classify the difference image based on one of a plurality of predefined classes; and
rendering the 3D ultrasound image of the tissue volume based on the series of 2D ultrasound images and the estimated distance.