Ultrasonic Probe Machine-Learned Image Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ultrasonic diagnostic apparatuses using plane wave transmission suffer from lower image quality due to increased mixing of reflected signals from other regions, compared to focused transmission.
Innovation Solution
An ultrasonic diagnostic apparatus that employs a machine-learned model to generate estimated image data equivalent to focused beam images from plane-wave beam image data, using a learning apparatus that performs machine learning with image data from both plane-wave and focused beam transmissions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If plane wave transmission is used to achieve high frame rate imaging, then imaging speed is improved, but image quality deteriorates due to increased mixing of reflected signals from other regions
Solution Approach 1:
A neural network model is introduced as an intermediary between the plane wave transmission data and the final image output. The neural network processes the raw plane wave transmission data and generates synthetic focused beam images, effectively mediating between the high-speed acquisition method and the high-quality image requirement without requiring actual focused beam transmission
Solution Approach 2:
The system creates a copy or replica of focused beam images by using a neural network trained on paired plane wave and focused beam image data. Instead of directly transmitting focused beams (which would reduce frame rate), the system synthesizes focused beam images from plane wave transmission data, producing a copy that mimics the quality of focused beam images while maintaining the speed advantage of plane wave transmission
2Measurement precision
If focused beam transmission is used to achieve high image quality, then image quality is improved, but frame rate decreases due to the need for multiple transmissions and receptions
Solution Approach 1:
The system generates synthetic focused beam images by processing plane wave transmission data through a neural network. This copying approach produces images with focused beam quality without requiring actual focused beam transmission, thereby maintaining the high frame rate of plane wave transmission while achieving the desired image quality
Solution Approach 2:
The system changes the processing parameters by applying a neural network model that transforms plane wave transmission data into focused beam images. This parameter transformation allows the system to operate in the faster plane wave mode while producing images with the quality characteristics of focused beam transmission through learned parameter mappings
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 solution enables high-quality images with improved contrast and resolution, comparable to focused beam transmission, while maintaining a high frame rate, thus addressing the image quality issues of plane wave transmission.
Implementation Method 1
an ultrasonic probe configured to transmit and receives ultrasonic waves to and from an observation region of an object
Implementation Method 2
reflected waves from a plane wave or a diffuse wave, transmitted in a plurality of directions or transmitted a plurality of times regarding a received signal
Implementation Method 3
an estimated image generating unit configured to generate estimated image data corresponding to image data based on an ultrasonic focused beam from image data obtained by transmission of an ultrasonic plane-wave beam by using a model having been machine-learned from learning data including image data obtained by the transmission of the ultrasonic plane-wave beam and image data obtained by the transmission of the ultrasonic focused beam
Data Source
AI summary
An ultrasonic diagnostic apparatus includes an ultrasonic probe configured to transmit and receives ultrasonic waves to and from an observation region of an object. The ultrasonic diagnostic apparatus further includes an estimated image generating unit configured to generate estimated image data corresponding to image data based on an ultrasonic focused beam from image data obtained by transmission of an ultrasonic plane-wave beam by using a model having been machine-learned from learning data including image data obtained by the transmission of the ultrasonic plane-wave beam and image data obtained by the transmission of the ultrasonic focused beam.