US 11,992,369 B2
Intelligent ultrasound system for detecting image artefacts
Raghavendra Srinivasa Naidu, Auburndale, MA (US); Man Nguyen, Melrose, MA (US); and Christine Menking Swisher, San Diego, CA (US)
Assigned to KONINKLIJKE PHILIPS N.V., Eindhoven (NL)
Appl. No. 16/760,962
Filed by KONINKLIJKE PHILIPS N.V., Eindhoven (NL)
PCT Filed Oct. 29, 2018, PCT No. PCT/EP2018/079529
§ 371(c)(1), (2) Date May 1, 2020,
PCT Pub. No. WO2019/086365, PCT Pub. Date May 9, 2019.
Claims priority of provisional application 62/580,635, filed on Nov. 2, 2017.
Prior Publication US 2020/0297318 A1, Sep. 24, 2020
This patent is subject to a terminal disclaimer.
Int. Cl. A61B 8/00 (2006.01); A61B 8/08 (2006.01); G06N 3/02 (2006.01)
CPC A61B 8/465 (2013.01) [A61B 8/5215 (2013.01); A61B 8/5269 (2013.01); G06N 3/02 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An ultrasound imaging system comprising:
an ultrasound transducer configured to acquire echo signals responsive to ultrasound pulses transmitted toward a target region;
one or more processors in communication with the ultrasound transducer and configured to:
generate at least one image frame from the acquired echo signals;
determine, via a neural network, a presence and type of an artifact in the at least one image frame from among different types of potential artifacts;
generate an indicator based on the determined presence of the artifact;
cause the indicator to be displayed on a user interface in communication with the processors;
generate an instruction for adjusting the ultrasound transducer based on the determined presence and type of the artifact, wherein the instruction includes a prompt to modify an operating parameter of the ultrasound transducer intended to remove the artifact from the at least one image frame, and wherein the operating parameter comprises a position, an orientation, a focal depth, a beam angle, an imaging modality, or a frequency of the ultrasound transducer; and
wherein the one or more processors are further configured to prepare the at least one image frame for input into the neural network by creating triplicate replications of the at least one image frame such that at least one of a plurality of neural network models can be utilized to determine the presence, an absence, and/or type of an artifact in the at least one image frame; and
wherein the triplicate replications of the at least one image frame includes: (i) a high-signal, low-dynamic range image; (ii) a low-signal, low-dynamic range image; and (iii) a compressed high-dynamic range image with an adaptive histogram equalization.