Registered Reference Images for Consistent Ultrasound Acquisition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge of achieving consistent ultrasound image acquisition across multiple time points, particularly in longitudinal examinations, is hindered by probe position drift and the need for reproducible image views to avoid measurement errors and ensure accurate clinical comparisons.
Innovation Solution
A method for ultrasound image acquisition guidance that utilizes a reference image from a datastore, registered with live ultrasound data to provide real-time guidance on adjusting the acquisition to match a predefined suitability criterion, using a suitability score and registration to align anatomical views and imaging settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual probe manipulation is used to acquire ultrasound views, then operator flexibility is maintained, but image consistency across time points deteriorates due to probe position drift
Solution Approach 1:
The system provides real-time feedback by comparing the current probe position with the reference position, displaying visual cues (e.g., alignment indicators, color-coded feedback) that guide the operator to adjust the probe until the current view matches the reference view, thereby achieving consistent image acquisition across time points
Solution Approach 2:
The patent replaces manual mechanical probe manipulation with an automated image-guidance system that uses computer vision and image processing algorithms to analyze ultrasound images, detect anatomical landmarks, and provide automated feedback for probe positioning, reducing reliance on operator skill and experience
2Productivity
If multiple clinicians perform measurements at different time points, then continuous monitoring is enabled, but measurement precision deteriorates due to varying acquisition techniques
Solution Approach 1:
The system standardizes acquisition parameters by storing reference images with specific imaging settings (gain, depth, focus, probe position) and guiding operators to replicate these parameters, ensuring that measurements taken by different clinicians at different time points are based on consistent imaging conditions
Solution Approach 2:
The system performs preliminary actions by pre-acquiring and storing high-quality reference images that serve as templates for subsequent acquisitions. These reference images are used to generate guidance feedback before the actual measurement is taken, ensuring that each clinician starts from the same standardized viewpoint
3Manufacturing precision
If probe position is adjusted to achieve correct views, then image quality improves, but time consumption increases due to repeated repositioning
Solution Approach 1:
The system provides real-time feedback during probe manipulation, showing the operator how close they are to achieving the correct view through visual indicators (e.g., alignment lines, color-coded feedback ranging from red to green). This immediate feedback eliminates the need for trial-and-error repositioning and reduces the time required to achieve accurate views
Solution Approach 2:
The system creates a digital copy of the reference image and overlays it with the current image, allowing the operator to visually compare and align anatomical structures. This image copying and overlay technique provides a direct visual template for achieving the correct probe position, significantly reducing adjustment time
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A method and system are provided for image acquisition guidance. The method is based on utilizing a reference image in a database and computing a registration offset between a live image view being acquired by an operator and the reference image. To improve consistency of image acquisition across different clinical applications, the method further comprises receiving an indication of a target image-based operation which is to be performed using the image being acquired by the operator and identifying a reference from a reference image store which meets a pre-defined suitability criterion for performing that operation. For example, this may comprise identifying an image in the store which has a highest value of a suitability score associated with suitability of the image for performing the given image based operation.