3D Ultrasound Panoramic Imaging With Adaptive Frame Patching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ultrasound imaging systems face challenges in acquiring high-resolution 3D images due to the use of expensive 2D-array transducers or mechanically moved 1D-array transducers, which have limited field of view and are prone to user-induced motion artifacts, and bone structures interfere with data acquisition under bone surfaces.
Innovation Solution
A guided technique for manually operating a 1D-array transducer using image registration to combine 2D ultrasound images into a 3D volume, with adaptive patching and smart patching to handle specular reflectors, reducing the need for expensive sensors and mechanical apparatuses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a 2D-array transducer is used to acquire 3D ultrasound volume, then the resolution and signal-to-noise ratio are improved, but the manufacturing cost increases significantly
Solution Approach 1:
The patent segments the 3D imaging task into multiple 2D image acquisitions taken at different positions and orientations. Instead of using a single complex 2D-array transducer, the system divides the volume acquisition into multiple sequential 2D scans that are later reconstructed into a 3D volume through image registration and stitching algorithms.
Solution Approach 2:
The patent uses multiple copies of simple 1D-array transducer images to reconstruct the 3D volume. Each 2D image serves as a copy of the anatomical structure from a specific viewpoint, and these copies are computationally assembled into a complete 3D representation, avoiding the need for a single expensive 2D-array transducer.
2Area of stationary object
If a mechanically moved 1D-array transducer is used to acquire 3D volume data, then the field of view is improved, but the device complexity and cost increase
Solution Approach 1:
The patent implements self-service by providing real-time guidance instructions to the user based on the current transducer position and the target examination area. The system automatically calculates and displays directional arrows and sweep patterns, allowing the user to self-correct their movement without external assistance or complex mechanical positioning mechanisms.
Solution Approach 2:
The patent replaces the mechanical moving apparatus with a software-based guidance system. Instead of using motors, rails, or automated positioning mechanisms to move the transducer, the system uses computer vision and image registration to track transducer position and provides visual feedback to guide manual movement, substituting mechanical automation with software intelligence.
3Device complexity
If manual movement of 1D-array transducer is used to acquire 3D volume, then the device complexity is reduced, but the image quality deteriorates due to inconsistent movement
Solution Approach 1:
The patent implements continuous feedback by displaying real-time guidance arrows and sweep patterns that indicate the correct movement direction and coverage area. The system monitors the current transducer position relative to the target anatomy and provides immediate visual feedback, allowing the user to adjust their movement in real-time to maintain consistent and complete coverage.
Solution Approach 2:
The patent performs preliminary action by pre-calculating the optimal sweep patterns and coverage areas before the user begins scanning. The system identifies the target anatomical boundaries and pre-generates guidance instructions that show the user exactly how to move the transducer to achieve complete and consistent coverage, eliminating the need for complex real-time correction mechanisms.
4Speed
If conventional image registration is used to combine 2D images into 3D volume, then the processing speed is maintained, but the image alignment precision deteriorates due to bone specular reflectors
Solution Approach 1:
The patent applies local quality by using different registration strategies for different anatomical regions. Instead of applying a single uniform registration algorithm to the entire volume, the system identifies regions with bone specular reflectors and applies specialized handling to those local areas, while using standard registration for soft tissue regions, thereby optimizing both speed and precision locally.
Solution Approach 2:
The patent changes parameters by adjusting the registration algorithm's sensitivity and weighting factors when bone structures are detected. The system dynamically modifies registration parameters such as correlation thresholds and transformation constraints based on the local tissue type, allowing faster processing in homogeneous soft tissue regions while maintaining higher precision in complex bone-containing areas.
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
Enables high-resolution 3D ultrasound imaging with reduced costs and improved image quality by guiding users through sweep patterns, minimizing gaps and artifacts, and effectively handling bone interference.
Implementation Method 1
ultrasound data generated by an ultrasound scanner based on reflections of ultrasound signals transmitted by the ultrasound scanner at an anatomy of a patient
Data Source
AI summary
Systems and methods for panoramic imaging in 2D and 3D ultrasound imaging are disclosed. The techniques disclosed herein use image registration to combine the 2D images into a 3D volume. Ultrasound images collected in different locations are combined using image tracking. When the transducer moves along a lateral direction (in line with a longitudinal axis of the transducer), a portion of the image is overlapped between frames and the overlapping frames can be patched. When the transducer moves along an elevational direction (non-parallel to the longitudinal axis of the transducer), there is no overlap between frames and the non-overlapping frames are not patched together but are stored side-by-side to create a volume. The storage and patching processes can be adaptive (e.g., the images are not stored or patched if there is minimal difference between neighboring frames).