Acquiring ultrasound image

Inventors

Putha, Preetham • Tadepalli, Manoj • Warier, Prashant • Rao, Pooja • Sahu, Rohan

Assignees

Qure AI Technologies Pvt Ltd

Abstract

A method and a system for acquiring a 3D ultrasound image. The method includes receiving a request to capture a plurality of ultrasound image for a medical test corresponding to a medical condition. The method further includes determining a body part corresponding to the medical test. Further, the method includes identifying an imaging site particular to the medical test. Furthermore, the method includes providing a navigational guidance to the user in real time for positioning a handheld ultrasound device. Subsequently, the user is assisted to capture the plurality of ultrasound image of the imaging site in real time using deep learning. Further, the plurality of ultrasound images of the imaging site is captured. Finally, the method includes converting the plurality of ultrasound image to a 3-Dimensional (3D) ultrasound image in real time.

Core Innovation

The invention relates to acquiring a 3-Dimensional (3D) ultrasound image for a medical test corresponding to a medical condition. A request to capture a plurality of ultrasound images is received from a user on a wireless display device, and based on a historical knowledge base, a body part corresponding to the medical test is determined and an imaging site is identified as a portion or predefined area of the body part related to the medical test.

The system provides navigational guidance in real time for positioning a handheld ultrasound device at the imaging site. The navigational guidance is based on deep learning and is provided in natural language, and the invention further assists capturing the plurality of ultrasound images in real time using deep learning, including real time feedback on at least image quality, automated measurements, coverage calculations, angular calculations, and characterizations specific to the ultrasound images of the medical condition.

The invention captures a plurality of ultrasound images as 2-Dimensional (2D) images when the handheld ultrasound device is positioned within a permissible radius of operation. The plurality of 2D ultrasound images is converted to a 3-Dimensional (3D) ultrasound image in real time by stitching the images together using an Artificial Intelligence (AI) enabled image model.

In dependent claim embodiments, the acquired 3D ultrasound image is analyzed with deep learning to identify predefined morphological characteristics of the imaging site in real time, compare them against a trained data model built from historical user data for the medical condition, and generate an analysis report. In further specialized embodiments, the imaging site and medical test are directed to carotid artery stenosis and include analysis using plaque-related and Doppler-related characteristics such as intima media thickness, plaque presence, plaque height, plaque density, doppler assessment, and plaque area.

Claims Coverage

The document includes three independent claim sets (method, system, and non-transitory computer program product). Across these independent claims, the inventive coverage centers on receiving a user request on a wireless display device, selecting a body part and imaging site from a historical knowledge base, providing deep-learning natural-language navigational guidance, assisting real-time capture with deep-learning feedback and condition-specific measurements/characterizations, capturing a set of 2D ultrasound images within a permissible radius, and converting the 2D images into a 3D ultrasound image using an AI-enabled image model.

The independent claims are directed to an AI-enabled workflow that uses a historical knowledge base to select a body part and imaging site, provides deep-learning natural-language real-time navigational guidance, and uses deep-learning assistance to capture condition-specific 2D ultrasound images that are stitched in real time with an AI-enabled image model into a 3D ultrasound image; dependent claim embodiments further add deep-learning extraction and comparison of predefined morphological characteristics and generation of an analysis report, including carotid stenosis characterization using plaque-related and Doppler-related characteristics.

Stated Advantages

Documented Applications