Iterative Medical Image Analysis Using Feedback and Blurring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing white matter hyperintensities in medical images, such as constrained energy minimization (CEM), face challenges including the inability to provide a threshold for classification, inability to analyze nonlinear images, and lack of spatial information, leading to analysis errors and inefficiencies.
Innovation Solution
An iterative analyzing method that uses image feedback and blurring to obtain spatial information, involving nonlinear dimensional expansion of spectral images, constrained energy minimization, and Gaussian filtering to detect lesions, with a predetermined similarity index for termination conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If constrained energy minimization (CEM) is used for analyzing brain MRI images, then the lesion can be detected to produce an abundance image, but it cannot provide a threshold for determining the classification and cannot analyze the area of white matter hyperintensities
Solution Approach 1:
The patent implements an iterative feedback mechanism where the abundance image from CEM is fed back through blurring and thresholding operations to generate a binary image, which then feeds back into the next iteration of CEM analysis. This closed-loop feedback system progressively refines the classification by repeatedly applying the analysis with updated spatial information, ultimately producing accurate classification thresholds for white matter hyperintensity detection.
2Measurement precision
If CEM is used as a linear filter for analyzing brain MRI images, then the abundance image can be produced, but it cannot analyze nonlinear images and cannot analyze the area of white matter hyperintensities
Solution Approach 1:
The patent introduces spatial dimensionality by applying blurring operations to the abundance image to create spatial spectral images. This transforms the purely spectral analysis into a spatio-spectral analysis, adding the spatial dimension that enables the method to handle nonlinear characteristics and accurately analyze white matter hyperintensity areas that linear filters alone cannot detect.
3Device complexity
If traditional hyperspectral image processing method only uses pixels as the basis for analysis, then the analysis can be simplified, but it does not include the spatial information of the image itself and will cause analyzing errors
Solution Approach 1:
The patent merges spectral information from CEM with spatial information from blurring operations to create spatial spectral images. This combination integrates both pixel-level spectral analysis and spatial context, allowing the method to maintain relatively simple implementation while significantly improving analysis accuracy by incorporating spatial relationships among pixels.
4Reliability
If manual visual evaluation and classification method is used for white matter hyperintensities, then the analysis can be performed, but it takes a lot of time to complete and increases the risk of analysis errors
Solution Approach 1:
The patent implements an automated iterative analysis system that performs white matter hyperintensity detection without manual intervention. The algorithm automatically executes multiple iterations of CEM, blurring, thresholding, and binary image generation, progressively refining the results until convergence. This self-service automated approach eliminates time-consuming manual operations while maintaining or improving analysis reliability through iterative refinement.
Data Source
AI summary
The present invention disclosed an iterative analyzing method, which can detect the lesion in the image quickly. In brief, the iterative analysis method of the medical image disclosed by the present invention is roughly as follows: first, the original spectral image cube is expanded into a spectral image cube by a method of nonlinear dimensional-expansion, and then detecting the target's subpixel by the method of constrained energy minimization to produce an abundance image; the abundance image is fed back to the spectral image cube for create another spectral image cube by the nonlinear method. Furthermore, the abundance image is only used for detecting the subpixel of the target and does not include any spatial information, so, in order to obtain the spatial information of spectral image, it obtains the spatial information around the subpixel by using a blurring tool such as a Gaussian filter. After the spatial information is fed back to the spectral image cube, the subpixel target detection is repeatedly performed until a predetermined termination condition is satisfied.