Medical Image Correction Using Temporary Records and Global Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing algorithms for correcting medical image data often result in local optima, failing to achieve global optima and leaving artifacts in the processed images due to high-dimensional optimization problems.
Innovation Solution
A method involving multiple temporary data records created by applying corrections with interference terms to image data, followed by a trained function to determine the global optimum based on image quality metrics, using neural networks for improved correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If image processing algorithms are configured as high-dimensional optimization problems, then correction capability is improved, but the risk of obtaining only local optima increases
Solution Approach 1:
The patent applies preliminary actions by generating multiple temporary data records with different interference terms before final selection. This allows the system to explore multiple optimization paths in advance, increasing the likelihood of finding the global optimum while maintaining high correction precision through systematic preliminary exploration.
Solution Approach 2:
The trained function serves as a feedback mechanism that evaluates multiple temporary data records and selects the optimal one. This feedback loop ensures that the system can identify whether it has found the global optimum or merely a local optimum, thereby improving optimization reliability while maintaining correction capability.
2Reliability
If multiple temporary data records are created and evaluated, then the likelihood of finding global optimum increases, but computational complexity increases
Solution Approach 1:
The patent applies parameter changes by systematically varying interference terms in the cost function across multiple temporary data records. This controlled parameter variation allows the system to explore the optimization landscape efficiently without requiring exhaustive search, thereby improving reliability of finding global optima while managing computational complexity through focused parameter exploration.
Solution Approach 2:
The system creates multiple copies of the data record with different interference terms applied. This copying approach allows parallel evaluation of different optimization paths without requiring fundamentally different computational structures, thereby increasing optimization reliability while keeping computational complexity manageable through replicated rather than fundamentally complex processing.
3Manufacturing precision
If a trained function is applied to determine global optimum, then correction accuracy is improved, but processing time increases
Solution Approach 1:
The trained function is applied as a preliminary evaluation step to quickly assess multiple temporary data records and identify the most promising candidate. This preliminary action reduces the need for extensive iterative optimization on all possible records, thereby improving correction accuracy while reducing overall processing time through early elimination of suboptimal candidates.
Solution Approach 2:
The system changes parameters by applying different interference terms to create varied temporary records, then uses the trained function to efficiently evaluate these variations. This parameter-based exploration combined with trained function evaluation improves correction accuracy by systematically exploring the solution space while managing processing time through focused rather than exhaustive evaluation.
Data Source
AI summary
A method includes receiving image data of an examination object. A first temporary data record is created by applying a first correction to the image data. A further temporary data record is created by applying a further correction to the image data. The further correction at least partially corresponds to the first correction. A trained function is applied to input data that is based on the first temporary data record and the further temporary data record. A parameter of the trained function is based on an image quality metric. It is determined whether the first temporary data record has a higher image quality compared with the further temporary data record. When a result is positive, the first temporary data record is provided as the corrected medical image data. When the result is negative, the further temporary data record is provided as the image data, and part of the method is repeated.