GPU Beamforming for High-Frame-Rate Ultrasonic Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high complexity and large data amounts in ultrafast ultrasonic imaging systems result in lower processing speeds, necessitating improved data processing methods to enhance imaging speed and efficiency.
Innovation Solution
An ultrasonic imaging device incorporating an analog-to-digital processing unit, buffer storage, imaging processing unit GPU, and image processing module to process and synthesize ultrasonic data in real-time, utilizing parallel processing and GPU computing for rapid image formation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If plane wave ultrasonic imaging is used to acquire data for the entire imaging area in one emission/reception event, then the imaging frame rate is greatly improved, but the complexity and data amount for back-end processing increases, resulting in lower imaging speed
Solution Approach 1:
The patent divides the ultrasonic signal processing into multiple stages: analog-to-digital conversion, beamforming, envelope detection, and image reconstruction. Each stage processes a portion of the data independently, allowing parallel computation and reducing the computational burden on any single processing unit. This segmentation enables the system to handle the large data volume from plane wave imaging while maintaining high frame rates.
Solution Approach 2:
The patent introduces a new dimension of processing by implementing a multi-dimensional data structure that organizes ultrasonic signals in terms of transducer elements, time delays, and spatial coordinates. This dimensional organization allows for efficient parallel processing across multiple GPUs and enables optimized memory access patterns, thereby reducing processing complexity while maintaining high imaging frame rates.
2Productivity
If plane wave ultrasonic imaging is used to acquire data for the entire imaging area in one emission/reception event, then the imaging frame rate is greatly improved, but the data amount for back-end processing becomes large, resulting in lower imaging speed
Solution Approach 1:
The patent extracts only the essential features from the raw ultrasonic data through beamforming and envelope detection before passing the data to the image reconstruction stage. By extracting key parameters such as signal amplitude, time of flight, and spatial position, the system reduces the data volume that needs to be processed while retaining all necessary information for high-quality image reconstruction, thus maintaining high frame rates.
Solution Approach 2:
The patent performs preliminary processing operations including analog-to-digital conversion, beamforming, and envelope detection before the main image reconstruction process. These preliminary actions prepare the data in an optimized format that reduces the computational load during the final image generation stage, enabling the system to handle large data amounts efficiently and maintain high imaging frame rates.
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 real-time calculation of key tissue parameters and improves ultrasonic imaging speed by synthesizing beams in real-time, achieving frame rates of up to 3000 frames per second.
Implementation Method 1
the plurality of frequency mixer circuits are separately configured to mix each analog radio frequency signal in the plurality of analog radio frequency signals, so as to obtain a plurality of first analog signals
Implementation Method 2
the plurality of filter circuits are separately configured to filter each first analog signal in the plurality of first analog signals
Implementation Method 3
the plurality of analog-to-digital conversion circuits are separately configured to perform analog-to-digital conversion processing for the plurality of second analog signals, so as to obtain a plurality of groups of digital IQ data
Implementation Method 4
the imaging processing unit GPU is configured to receive the plurality of groups of digital IQ data from a buffer storage unit, and to at least partially perform the imaging processing in parallel for the plurality of groups of digital IQ data
Data Source
AI summary
Provided are ultrasonic imaging device and ultrasonic imaging system. Ultrasonic imaging device includes analog-to-digital processing unit, buffer storage unit, imaging processing unit GPU, and image processing module. Analog-to-digital processing unit includes first interface, multiple frequency mixer circuits, multiple filter circuits, multiple analog-to-digital conversion circuits, and second interface. First interface receives, in parallel, multiple analog radio frequency signals formed by ultrasonic waves sensed and returned by multiple sensors of probe. Second interface outputs multiple groups of digital IQ data. Buffer storage unit receives, buffers, and stores digital IQ data. Imaging processing unit GPU, at least in part and in parallel, performs imaging processing on digital IQ data, to respectively form multiple raw image data of multiple pixel points of multiple image lines in multiple image rows of image. Image processing module forms image data of ultrasonic imaging based on raw image data.