Systems and methods for volume-based ultrasound guidance and image analysis
AssigneeCaption Health, Inc.
InventorsHa Hong, Kilian Koepsell, Michael G. Cannon, Nripesh Parajuli, Nicolas Poilvert
Life of this patent
Estimating the term…
Abstract
Provided herein are platforms, systems, and methods for providing ultrasound imaging guidance for automatic evaluation of image quality over a plurality of sub-views.
Claims20
A method for guiding an ultrasound imaging procedure, comprising:
- receiving one or more ultrasound images of a subject acquired through an ultrasound imaging system;
- evaluating said one or more ultrasound images using a trained machine learning model comprising a representation of a three-dimensional structure of a target volume of the subject to determine a current position of a probe of the ultrasound imaging system and/or a clinical image quality;
- identifying an optimal probe position for imaging for at least one sub-view of the target volume of the subject;
- generating one or more probe movements for reducing a difference between the current probe position and the identified optimal probe position;
- and
- providing guidance instructions to a user of the ultrasound imaging system to make at least one generated probe movement.
An ultrasound system comprising at least one processor, a memory, and an ultrasound imaging probe, the memory comprising computer executable instructions that, when executed by the at least one processor, causes the processor to:
- receive one or more ultrasound images of a subject acquired through the ultrasound imaging probe;
- evaluate said one or more ultrasound images using a trained machine learning model comprising a representation of a three-dimensional structure of a target volume of the subject to determine a current position of a probe of the ultrasound imaging system and/or a clinical image quality;
- identify an optimal probe position for imaging for at least one sub-view of the target volume of the subject;
- generate one or more probe movements for reducing a difference between the current probe position and the identified optimal probe position;
- and
- provide guidance instructions to a user of the ultrasound imaging system to make at least one generated probe movement.
A non-transitory computer-readable storage media encoded with instructions that, when executed by a processor, causes the processor to:
- receive one or more ultrasound images of a subject acquired through an ultrasound imaging probe of an ultrasound imaging system;
- evaluate said one or more ultrasound images using a trained machine learning model comprising a representation of a three-dimensional structure of a target volume of the subject to determine a current position of a probe of the ultrasound imaging system and/or a clinical image quality;
- identify an optimal probe position for imaging for at least one sub-view of the target volume of the subject;
- generate one or more probe movements for reducing a difference between the current probe position and the identified optimal probe position;
- and
- provide guidance instructions to a user of the ultrasound imaging system to make at least one generated probe movement.
Dependencies are inferred from the claim text.