OCT Data Geometric Correction via Morphological Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Optical coherence tomography (OCT) data is distorted by sample movement during scanning, leading to geometric transformations such as displacement, which can result in incorrect diagnosis due to misinterpretation of 3D structures, especially in ophthalmology where patient movement causes saccade movements, and existing correction methods require additional measurements or are solely post-processing based.
Innovation Solution
A method for geometric correction of OCT data that involves obtaining scan morphological data, comparing it with reference morphological data to determine relative geometric transformations, and generating a corrected scan by applying a transform based on these transformations, which can be iteratively processed in both directions to account for displacements in the z-axis and potentially the x-axis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If additional tomograms are captured in y-axis direction for correction, then correction accuracy is improved, but acquisition time increases
Solution Approach 1:
The patent extracts only the essential morphological data (layer boundary positions) from the OCT scans that are needed for correction, rather than using additional complete tomograms. This allows correction to be performed using minimal additional information, reducing acquisition time while maintaining correction accuracy.
Solution Approach 2:
The patent performs preliminary detection of layer boundaries and morphological features during the standard scan acquisition. This preliminary extraction of correction data allows the actual correction to be done without requiring additional full tomogram captures, thus improving correction accuracy without increasing overall acquisition time.
2Loss of time
If post processing correction only is used, then acquisition time is reduced, but correction accuracy deteriorates
Solution Approach 1:
The patent performs preliminary detection of layer boundaries and morphological features during the standard scan acquisition phase. This preliminary action captures essential correction data without requiring additional acquisition time, while still enabling accurate post-processing correction through iterative optimization algorithms.
Solution Approach 2:
The patent implements feedback through iterative optimization where the correction algorithm continuously refines displacement estimates by comparing corrected images against expected morphological consistency. This feedback mechanism achieves high correction accuracy using only the originally acquired data without additional measurements.
3Productivity
If intensity profile correlation is used for correction, then correction speed is improved, but correction accuracy deteriorates due to intensity variations
Solution Approach 1:
The patent changes the parameter basis for correction from intensity profiles to morphological parameters (layer boundary positions). This parameter change makes the correction immune to intensity variations caused by illumination or tissue properties, maintaining both speed and accuracy by working with geometric rather than photometric data.
Solution Approach 2:
The patent substitutes the mechanical/intensity-based correlation method with a morphology-based geometric alignment approach. By replacing intensity profile correlation with layer boundary position matching, the system achieves correction that is independent of intensity variations while maintaining computational efficiency.
4Measurement precision
If morphological data comparison is performed, then correction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the complex task of full-image morphological comparison into focused comparisons of specific layer boundaries and key morphological features. This segmentation reduces computational complexity by concentrating processing on critical anatomical landmarks rather than performing exhaustive pixel-by-pixel or region-by-region comparisons across entire images.
Solution Approach 2:
The patent applies local quality by focusing morphological comparison on specific critical regions (layer boundaries and key anatomical landmarks) rather than uniformly processing the entire image. This localized approach maintains high correction accuracy at critical interfaces while reducing overall computational burden by ignoring less critical regions.
Data Source
AI summary
System and method for geometric correction of OCT data representing a scan obtained by means of optical coherence tomography imaging of a sample. First scan morphological data relating to a morphology of the sample in the scan are obtained. Then reference morphological data relating to a reference morphology of the sample are obtained. For each of one or more scans of OCT data scan morphological data are compared with the reference morphological data to determine a relative geometric transformation of the morphology of the scan with respect to the reference morphology. Corrected scan is generated by performing on the OCT data representing the scan a transform that relates to the determined relative geometric transformation.