Nonlinear Waveform Inversion with Measured-Output Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ultrasound imaging methods, such as B-mode imaging and inverse ultrasound algorithms, fail to provide sufficient contrast and accurate reconstruction of physical properties like speed-of-sound, density, and elasticity in non-linear media, and existing neural network approaches require extensive training data and lack interpretability.
Innovation Solution
A nonlinear waveform inversion system using a neural network platform models wave propagation in non-linear media, employing a wave field modeler, properties adjuster, and medium properties recoverer to optimize and recover physical properties through a backpropagation process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full waveform inversion algorithm is used to reconstruct physical properties, then measurement precision is improved, but productivity deteriorates due to computationally heavy iterative gradient-based approach
Solution Approach 1:
The patent replaces the traditional iterative gradient-based computational mechanics of full waveform inversion with a neural network-based system. The neural network is trained offline to learn the mapping from ultrasound signals to physical properties, and during operation, it directly predicts properties without iterative computation, substituting the heavy mechanical optimization process with a trained inference model that provides both accuracy and speed.
2Productivity
If machine learning approaches are used to estimate speed-of-sound, then productivity is improved, but measurement precision deteriorates due to poor interpretability and data dependency
Solution Approach 1:
The patent incorporates a feedback mechanism where the neural network's predictions are refined through a loss function that compares predicted transducer outputs with actual measured outputs. The properties adjuster uses backpropagation to optimize the physical properties estimates, creating a feedback loop that ensures both speed (through neural network inference) and precision (through iterative refinement based on measurement errors).
Solution Approach 2:
The system segments the problem into distinct functional modules: wave field modeler (generating predicted signals), properties adjuster (optimizing properties through loss function minimization), and medium properties recoverer (outputting final recovered properties). This segmentation allows each component to specialize, with the neural network handling speed and the optimization loop handling precision, resolving the contradiction between computational efficiency and measurement accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively reconstructs physical properties with improved contrast and resolution, including nonlinearity, in non-linear media, overcoming computational inefficiencies and data dependency of previous methods.
Implementation Method 1
a neural network having a neural network representation of a non-linear wave function of a set of physical properties of the wave field
Implementation Method 2
The properties adjuster operates a backpropagator using the neural network representation of the wave function and activates the wave field modeler with the improved set of the physical properties
Data Source
AI summary
A system for recovering physical properties from a non-linear medium includes a wave field modeler, a properties adjuster and a medium properties recoverer. The wave field modeler models a wave field generated by a transmitted pulse traveling in the non-linear medium and generates a predicted transducer output from the modeled wave field. The wave field modeler is a neural network having a neural network representation of a non-linear wave function of the physical properties of the wave field. The properties adjuster optimizes a loss function between the predicted transducer output and a measured transducer output and generates an improved set of the physical properties. The properties adjuster operates a backpropagator using the neural network representation and activates the wave field modeler with the improved set of the physical properties. The medium properties recoverer outputs a current improved set of the physical properties once the properties adjuster finishes operation.