Ultrasound Needle Position Detection via Candidate Region Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ultrasound diagnosis apparatuses face challenges in clearly visualizing puncture needles due to scattering of ultrasound waves, leading to inaccurate detection of needle positions, especially near the edge, which can result in incorrect information and additional processing requirements.
Innovation Solution
An ultrasound diagnosis apparatus that includes a candidate region extracting unit, a needle identification range setting unit, and a needle emphasizing unit to determine and emphasize the estimated range of the puncture needle position based on the aggregation state of candidate detection regions, using parameters like the number and variance of candidate regions, and weighting to enhance visualization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ultrasound waves are used to image a puncture needle, then the needle can be visualized in the ultrasound image, but the needle edge is not clearly shown due to scattering of ultrasound waves
Solution Approach 1:
The patent segments the needle detection problem into multiple components: extracting candidate regions, determining needle identification range, estimating needle position range based on aggregation state, and emphasizing the estimated range. This segmentation allows each component to be optimized independently, improving overall detection reliability while managing the precision challenge through systematic processing stages
Solution Approach 2:
The patent transitions from direct edge detection in the image space to analyzing the aggregation state of candidate regions across multiple processing dimensions. By evaluating candidate region distribution patterns and aggregation metrics, the system determines needle position in a transformed parameter space, then maps back to the original image space for emphasis, effectively navigating around the scattering problem
2Measurement precision
If edge detection is performed based on luminance distribution, then the needle position can be identified, but wrong positions are detected due to noise or other reflecting structures
Solution Approach 1:
The patent performs preliminary extraction of candidate regions and determination of needle identification range before final position estimation. By pre-processing the image data to identify potential needle regions and establishing the identification range based on aggregation state, the system prepares filtered and organized data that reduces the impact of noise and false reflectors in the subsequent detection stage
Solution Approach 2:
The patent incorporates feedback mechanisms by using the aggregation state of candidate regions to adjust and refine the estimated needle position range. The system evaluates the distribution and concentration of candidate regions, uses this aggregation information to determine the most likely needle position, and applies emphasis accordingly, creating a feedback loop that improves detection reliability by validating results against expected aggregation patterns
3Measurement precision
If additional structures and image processing are added to identify needle edges, then the needle position can be detected, but the device complexity and processing requirements increase
Solution Approach 1:
The patent merges multiple functions into a unified processing framework: candidate region extraction, needle identification range determination, position estimation based on aggregation state, and emphasis processing are integrated into a cohesive workflow. This merging reduces the need for separate, complex structures by combining detection and emphasis functions into a streamlined process that achieves precise needle edge identification through coordinated operation of integrated components
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 allows for more accurate and reliable visualization of puncture needle positions, reducing the likelihood of incorrect detections and simplifying the identification process by emphasizing the most likely range of the needle within the ultrasound image.
Implementation Method 1
an ultrasound probe 2 that emits ultrasound waves into a test object and receives waves (echo) reflected from the inside of the test object
Implementation Method 2
Particularly, in the vicinity of the edge of the puncture needle, ultrasound waves are easily scattered
Data Source
AI summary
An ultrasound diagnosis apparatus generating an ultrasound image of inside of a test object based on ultrasound signals reflected and received from the inside of the test object includes: a candidate region extracting unit which extracts a plurality of candidate detection regions in a needle position range corresponding to a position of a puncture needle inserted into the test object; a needle identification range setting unit which determines a needle identification range including the needle position range from the ultrasound image based on the extracted candidate detection regions; a needle position range estimating unit which determines an estimated range of the needle position range in the needle identification range based on a value indicating an aggregation state related to a distribution of the candidate detection regions in the needle identification range; and a needle emphasizing unit which performs a process for emphasizing the estimated range in the ultrasound image.