OCT Image Processing With Robust Weights for Deep-Layer Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing OCT image processing techniques fail to adequately reduce projection artifacts in deep layer images due to the influence of outliers in shallow and deep layer images, leading to increased correlation and artifacts in vascular structures.
Innovation Solution
An OCT image processing method using a robust estimation approach, such as weighted regression, to calculate correction weights that minimize the correlation between shallow and deep layer images, effectively reducing the impact of projection artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional weight calculation methods are used to reduce correlation between shallow and deep layer images, then projection artifacts are reduced to some extent, but outliers in the images prevent appropriate weight calculation and projection artifacts are not sufficiently reduced
Solution Approach 1:
The patent changes the parameter calculation method from conventional correlation-based weight calculation to robust estimation-based weight calculation. This parameter change allows the system to handle outliers effectively by using statistical methods that are less sensitive to extreme values, thereby improving both the accuracy of weight calculation and the reduction of projection artifacts.
Solution Approach 2:
The patent replaces the conventional mechanical correlation calculation method with a robust estimation method that uses statistical principles. This substitution enables the system to calculate weights more accurately in the presence of outliers by relying on statistical properties rather than direct correlation measurements, which are easily affected by outlier values.
2Object-affected harmful factors
If correlation reduction between shallow and deep layer images is prioritized, then projection artifacts are reduced, but the presence of outliers prevents appropriate correction weight calculation
Solution Approach 1:
The patent changes the parameter estimation approach from conventional methods to robust estimation methods. This parameter change improves the reliability of correction weight calculation by using statistical techniques that are designed to be insensitive to outliers, thereby ensuring reliable weight calculation even when the data contains anomalous values.
Solution Approach 2:
The patent implements a feedback mechanism where the robust estimation method continuously adjusts the weight calculation based on the statistical properties of the data. This feedback loop allows the system to identify and compensate for the presence of outliers, improving the reliability of the correction weight calculation by iteratively refining the weights based on statistical feedback from the data distribution.
Data Source
AI summary
An OCT image processing program causes an OCT image processing device to perform: an image acquisition step of acquiring a shallow layer image and a deep layer image which are generated based on the motion contrast data, wherein the shallow layer image is an image of a shallow region of the living tissue, and the deep layer image is an image of a deep region deeper than the shallow region; a correction weight calculation step of calculating a correction weight for correcting the deep layer image such that a correlation between the shallow layer image and the deep layer image is reduced; and an image correction step of correcting the deep layer image according to the calculated correction weight. At the correction weight calculation step, the correction weight is calculated using a robust estimation method.