US 11,813,113 B2
Automated extraction of echocardiograph measurements from medical images
Ehsan Dehghan Marvast, Palo Alto, CA (US); Allen Lu, Bellevue, WA (US); and Tanveer F. Syeda-Mahmood, Cupertino, CA (US)
Filed by Merative US L.P., Ann Arbor, MI (US)
Filed on Mar. 18, 2021, as Appl. No. 17/205,485.
Application 17/205,485 is a continuation of application No. 16/713,999, filed on Dec. 13, 2019, granted, now 10,987,013.
Application 16/713,999 is a continuation of application No. 15/848,077, filed on Dec. 20, 2017, granted, now 10,531,807, issued on Jan. 14, 2020.
Prior Publication US 2021/0204856 A1, Jul. 8, 2021
Int. Cl. G06T 7/00 (2017.01); G06T 7/62 (2017.01); A61B 8/08 (2006.01); A61B 5/00 (2006.01); G16H 30/40 (2018.01); A61B 5/318 (2021.01); A61B 5/316 (2021.01); G06V 10/44 (2022.01); G06F 18/21 (2023.01); G06F 18/25 (2023.01); G06V 10/80 (2022.01); G16H 10/60 (2018.01)
CPC A61B 8/0883 (2013.01) [A61B 5/316 (2021.01); A61B 5/318 (2021.01); A61B 5/72 (2013.01); A61B 8/5223 (2013.01); G06F 18/21 (2023.01); G06F 18/253 (2023.01); G06T 7/0012 (2013.01); G06T 7/0014 (2013.01); G06T 7/62 (2017.01); G06V 10/454 (2022.01); G06V 10/806 (2022.01); G16H 30/40 (2018.01); G06T 2207/10132 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30048 (2013.01); G06V 2201/031 (2022.01); G16H 10/60 (2018.01)] 20 Claims
OG exemplary drawing
 
1. A method, in a data processing system comprising at least one processor and at least one memory, the at least one memory comprising instructions that are executed by the at least one processor to cause the at least one processor to implement an automated echocardiograph measurement extraction system, the method comprising:
training a deep learning network, via a machine learning process based on an annotated training medical image data set, having training medical images of different viewpoints and corresponding ground truth data specifying ground truth echocardiograph measurements, for each type, of one or more types, of echocardiograph measurements required to be extracted from medical images, to learn a corresponding medical image viewpoint that provides an optimum viewpoint for generating the type of echocardiograph measurement, thereby generating a trained deep learning network;
receiving, by the automated echocardiograph measurement extraction system executing on the data processing system, medical imaging data comprising one or more medical images;
inputting, by the automated echocardiograph measurement extraction system, the one or more medical images into the trained deep learning network;
automatically processing, by the trained deep learning network, the one or more medical images to generate an extracted echocardiograph measurement vector output comprising one or more values for each of the one or more types of echocardiograph measurements extracted from the one or more medical images, wherein the echocardiograph measurements are measurements of dimensions of anatomical structures, and wherein a value for each type of echocardiograph measurement is extracted from at least one selected medical image in the one or more medical images selected based on the training of the deep learning network to learn the corresponding medical image viewpoint that provides the optimum viewpoint; and
outputting, by the trained deep learning network, the extracted echocardiograph measurement vector output to a medical image viewer.