Flexible Ultrasound Array Shape Estimation Without External Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ultrasound imaging systems face challenges in accurately determining the spatial configuration of flexible or stretchable transducer arrays without relying on external tracking systems, leading to increased cost, complexity, and potential errors in imaging artifacts, particularly in complex or dynamically changing geometries.
Innovation Solution
A method and system that utilizes machine learning-based processing of ultrasound data features, such as time-of-flight, amplitude, and cross-correlation, to estimate the spatial configuration of transducer elements without external tracking systems, incorporating a DNN trained on synthetic and real data to infer precise coordinates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If external tracking systems (optical tracking, optical fibers, strain sensors) are used to determine spatial configuration of flexible transducer elements, then localization accuracy is improved, but system cost and complexity increase
Solution Approach 1:
The patent extracts and processes ultrasound signal features (time-of-flight, amplitude, cross-correlation) directly from the ultrasound data itself, eliminating the need for external tracking systems. The machine learning model processes these extracted features to determine transducer element positions, achieving localization without additional hardware components.
Solution Approach 2:
The ultrasound system uses its own acquired data (ultrasound signals) to determine the spatial configuration of the transducer array. The machine learning model processes features extracted from the ultrasound data itself to provide localization information, making the system self-sufficient without requiring external tracking infrastructure.
2Measurement precision
If iterative optimization methods are used to estimate array shape through optimization of image quality metrics, then localization accuracy is improved, but computational intensity increases
Solution Approach 1:
The patent replaces iterative optimization algorithms with a machine learning-based approach. The machine learning model processes extracted ultrasound features to directly estimate transducer positions without requiring iterative optimization loops, significantly reducing computational intensity while maintaining accuracy.
Solution Approach 2:
The patent changes the approach from optimizing image quality metrics iteratively to using machine learning models that process specific extracted features (time-of-flight, amplitude, cross-correlation). This parameter transformation enables direct estimation without iterative computation, reducing energy consumption while maintaining localization precision.
3Reliability
If traditional fixed arrays of transducer elements are used, then imaging reliability is maintained, but adaptability to complex geometries is reduced
Solution Approach 1:
The patent employs flexible or stretchable transducer arrays that can dynamically adapt their configuration to complex or dynamically changing geometries. The machine learning-based localization system processes ultrasound data to determine the actual spatial configuration of these dynamic arrays, enabling reliable imaging across various geometries without requiring rigid fixed structures.
4Adaptability or versatility
If flexible or stretchable transducer arrays are used to adapt to complex geometries, then adaptability is improved, but imaging artifacts increase due to misalignment
Solution Approach 1:
The patent implements a feedback mechanism where the machine learning model processes ultrasound data to continuously determine the actual spatial configuration of the flexible transducer array. This feedback information is used to correct for misalignment and update the spatial mapping, ensuring accurate image reconstruction even as the array deforms to adapt to complex geometries.
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
This approach reduces system cost and complexity while improving accuracy and generalizability, enabling high-resolution imaging in scenarios involving non-planar surfaces or dynamically moving objects, particularly beneficial in medical and industrial applications.
Implementation Method 1
images are created by transmitting soundwaves into the object under investigation from one or multiple elements of a transducer array and subsequently recording the echoes reflected by the interior of the object
Implementation Method 2
A method and system that utilizes machine learning-based processing of ultrasound data features, such as time-of-flight, amplitude, and cross-correlation, to estimate the spatial configuration of transducer elements
Data Source
AI summary
A method for locating an unknown arrangement of ultrasound transducer elements includes acquiring a set of raw ultrasound data using a flexible or rigid array of transducer elements, extracting a set of features from the ultrasound data, the features being derived from fundamental physics of ultrasound propagation and structural characteristics of a target object being imaged, organizing and structuring the extracted features as input for a machine learning model, and determining spatial coordinates of each ultrasound transducer element based on an output of the machine learning model.