Method and system for generating a synthetic elastography image

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional shear-wave elastography techniques require specialized ultrasound probes and high computational resources, making them unavailable in many ultrasound systems and computationally demanding.

Innovation Solution

A method using a trained artificial neural network to generate synthetic elastography images from conventional B-mode ultrasound images, leveraging texture information to produce elasticity maps without the need for specialized hardware or ultrafast imaging.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If shear-wave elastography is implemented using conventional techniques, then elastography images can be generated, but specialized ultrasound probes and high computational resources are required

Engineering Contradiction:

Improveavailability of elastography imagingVSAvoidspecialized hardware requirements

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent uses a trained artificial neural network to generate synthetic elastography images that replicate the appearance and diagnostic information of true elastography images, but derived from conventional B-mode ultrasound images instead of requiring specialized elastography acquisition sequences

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical/acoustic push-pulse system with a computational system that uses machine learning to infer elasticity information from standard B-mode images, eliminating the need for specialized transducers and ultrafast imaging capabilities

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If shear-wave elastography is performed with ultrafast imaging at 1000 Hz frame rate, then accurate shear-wave speed measurement is achieved, but computational demand increases and frame rate decreases

Engineering Contradiction:

Improveshear-wave speed measurement accuracyVSAvoidimaging frame rate

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the ultrafast mechanical imaging system with a computational approach using a trained neural network that processes conventional B-mode images to directly generate elastography images, eliminating the need for ultrafast frame rates while maintaining diagnostic accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The neural network is pre-trained on a large dataset of paired B-mode and elastography images, allowing it to perform the complex computational task of elasticity estimation during the preprocessing training phase, enabling rapid inference during actual imaging without requiring ultrafast acquisition

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12369887B2Method and system for generating a synthetic elastography image

Publication Date: 2025.07.29 KONINKLIJKE PHILIPS NV

AI summary

The invention relates to a method for generating a synthetic elastography image (18), the method comprising the steps of (a) receiving a B-mode ultrasound image (5) of a region of interest; (b) generating a synthetic elastography image (18) of the region of interest by applying a trained artificial neural network (16) to the B-mode ultrasound image (5). The invention also relates to a method for training an artificial neural network (16)5 useful in generating synthetic elastography images, and a related computer program and system.