Super-Resolution Medical Imaging via Progressive Sub-Voxel Up-Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical images often have fixed resolutions that balance acquisition time with image quality, but higher resolution images are desired for easier identification of anatomical features for diagnosis and treatment.
Innovation Solution
A method for progressive sub-voxel up-sampling of medical images using a deep neural network to generate intermediate images with higher resolutions, ultimately producing a super-resolution output image with increased clarity and interpretability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image resolution is increased to improve anatomical feature identification, then diagnostic accuracy is improved, but acquisition time increases
Solution Approach 1:
The system performs up-sampling of lower-resolution images to generate higher-resolution images in advance, allowing diagnostic personnel to work with high-resolution images without requiring long acquisition times. The up-sampled images are generated before the actual diagnostic process, enabling faster workflow.
Solution Approach 2:
The system creates copies of lower-resolution images at higher resolutions through up-sampling algorithms. These copied high-resolution images serve as substitutes for actual high-resolution acquisitions, providing the necessary detail for diagnosis without the time penalty of acquiring true high-resolution images.
2Measurement precision
If image resolution is increased to improve anatomical feature identification, then diagnostic accuracy is improved, but image acquisition complexity increases
Solution Approach 1:
The system uses computational copying through up-sampling algorithms to generate high-resolution images from lower-resolution inputs. This approach avoids the need for complex hardware configurations required for direct high-resolution acquisition, simplifying the overall system while achieving the desired image quality.
Solution Approach 2:
The system replaces mechanical/acquisition-based resolution improvement with computational processing. Instead of using complex imaging hardware to directly capture high-resolution images, the system uses software-based up-sampling algorithms to achieve the same diagnostic goal with simpler acquisition protocols.
Data Source
AI summary
Various methods and systems are provided for generating super-resolution images. In one embodiment, a method comprises: progressively up-sampling an input image to generate a super-resolution output image by: generating N intermediate images based on the input image, where N is equal to at least one, including a first intermediate image by providing the input image to a deep neural network, where a resolution of the first intermediate image is a multiple of a resolution of the input image, higher than the resolution of the input image, and can be any positive real value and not necessarily an integer value; generating the super-resolution output image based on the N intermediate images, the super-resolution output image having a resolution higher than a respective resolution of each intermediate image of the N intermediate images and the resolution of the input image; and displaying the super-resolution output image via a display device.