Retinal Layer Segmentation in OCT Angiography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical coherence tomography (OCT) methods are limited in detecting blood flow abnormalities such as capillary dropout or pathologic vessel growth, which are key features of age-related macular degeneration and proliferative diabetic retinopathy, due to their inability to directly detect blood flow or discriminate vascular tissue from surrounding tissue, and require invasive dye-based contrast agents.
Innovation Solution
The development of methods and systems for automated and manual segmentation of retinal layers in OCT scans, using directional graph search and manual editing techniques, to separate and visualize blood flow data into specific layers, facilitating analysis and visualization of ocular abnormalities like retinal and choroidal neovascularization, and macular edema.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual delineation of retina layer boundaries is performed by an experienced expert, then segmentation accuracy is improved, but time consumption and operator dependency increase
Solution Approach 1:
The patent segments the retinal image processing task into multiple components: automated boundary detection using image processing algorithms, extraction of layer boundary information, and optional manual refinement. This segmentation allows the system to perform initial accurate segmentation automatically, reducing time consumption while maintaining the option for expert review when needed.
Solution Approach 2:
The system implements self-service through automated segmentation algorithms that can independently process retinal images without requiring expert intervention for every case. The algorithm automatically detects layer boundaries, extracts features, and generates segmentation results, freeing experts from routine tasks while maintaining high accuracy through algorithmic consistency.
2Productivity
If automated segmentation algorithms are used, then time efficiency is improved, but segmentation accuracy deteriorates in datasets with pathologies such as drusen, cystoid macular edema, subretinal fluid or pigment epithelial detachment
Solution Approach 1:
The patent implements a dynamic segmentation system that adapts its approach based on the characteristics of the input image. The algorithm automatically adjusts its parameters and processing steps according to the detected pathology type and severity, allowing it to maintain high accuracy across diverse pathological conditions while preserving time efficiency through automation.
Solution Approach 2:
The system incorporates feedback mechanisms where segmentation results are evaluated and used to refine subsequent processing steps. When pathologies are detected, the system adjusts its segmentation strategy based on feedback from intermediate processing stages, allowing it to correct for distortion effects and maintain accuracy in challenging cases.
3Speed
If conventional structural OCT is used, then imaging speed is improved, but the ability to detect blood flow abnormalities deteriorates
Solution Approach 1:
The patent merges structural OCT imaging with blood flow detection capabilities into a unified system. By combining the fast imaging speed of structural OCT with additional processing algorithms that extract blood flow information from the same dataset, the system achieves both rapid imaging and enhanced detection of vascular abnormalities without requiring separate imaging modalities.
Solution Approach 2:
The system achieves multi-functionality by enabling a single OCT device to perform both structural imaging and blood flow detection. The same optical coherence tomography system can generate structural images for morphological assessment and simultaneously process the data to detect blood flow abnormalities, eliminating the need for separate imaging procedures and contrast agents.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate and efficient segmentation of retinal layers, allowing for qualitative and quantitative assessment of blood flow and structural changes, reducing the need for invasive contrast agents and improving diagnostic accuracy in retinal pathologies.
Implementation Method 1
Optical coherence tomography (OCT) is a noninvasive, depth resolved, volumetric imaging technique that provides cross-sectional and three-dimensional (3D) imaging of biological tissues
Implementation Method 2
A limitation of conventional structural OCT, however, is that it is only sensitive to backscattered light intensity
Implementation Method 3
OCT angiography is a refinement of the OCT imaging technique that uses the motion of red blood cells against static tissue as intrinsic contrast to allow visualization of blood flow
Data Source
AI summary
Disclosed herein are methods and systems for segmenting, visualizing, and quantifying the layered structure of retina in optical coherence tomography datasets. The disclosed methods have particular application to OCT angiography data, where specific retina layers have distinct vascular structures and characteristics that can be altered in various pathological conditions of the eye.