Ultrasound Image Rendering Guided by Feature Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ultrasound image transformation techniques often fail to produce high-quality images due to the generation of abnormal results.
Innovation Solution
A computer-implemented method that derives classification information for features in ultrasound images and uses machine learning models, such as generative adversarial networks and diffusion models, to condition the image transformation process, reducing the generation of abnormal images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If image transformation techniques are applied to ultrasound images, then the visual quality and photo-realism of rendered images is improved, but the reliability and accuracy of the transformation process deteriorates due to generation of abnormal results
Solution Approach 1:
The patent segments the image transformation process into distinct stages: obtaining the input ultrasound image, deriving classification information for different features, and then performing the transformation using machine learning models conditioned on this classification information. This segmentation allows each stage to be optimized independently, improving both visual quality and transformation reliability.
Solution Approach 2:
The patent performs preliminary action by deriving classification information for different features in the ultrasound image before applying the image transformation. This preliminary classification step provides conditioning information that guides the subsequent transformation process, preventing abnormal results and improving transformation accuracy.
2Reliability
If classification information is derived for multiple features, then the reliability of image transformation is improved, but the device complexity and processing requirements increase
Solution Approach 1:
The patent employs machine learning models that can perform multiple functions: they process the input ultrasound image, derive classification information for various features, and execute the image transformation all within a unified system. This multi-functionality reduces overall system complexity while maintaining high transformation reliability.
Solution Approach 2:
The patent introduces classification information as an intermediary element that mediates between the input ultrasound image and the transformation process. This intermediary classification layer simplifies the overall processing architecture by providing structured conditioning information that guides the transformation without requiring complex direct processing.
Data Source
AI summary
A method for processing ultrasound imaging data comprising: obtaining a two-dimensional ultrasound image; deriving from input data, classification information for each of a plurality of features in the two-dimensional ultrasound image, the input data comprising at least one of: the two-dimensional ultrasound image or three-dimensional ultrasound data corresponding to the two-dimensional image; deriving a rendered image by supplying to an image transformation machine learning model: the two-dimensional ultrasound image as an input image; and the classification information for each of the plurality of features. The classification information provides, for example, classification of different body parts of an imaged subject that can be used to condition the image transformation process to reduce the generation of abnormal images.