Automated Fovea Selection in Optical Coherence Tomography Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Optical Coherence Tomography (OCT) systems face challenges in efficiently selecting clinically relevant images, particularly those containing the fovea, which are crucial for retinal disease diagnosis, due to the high volume of data and variability in image quality.
Innovation Solution
A method and system that utilize derivative calculations, Bayesian level set algorithms, and probability distributions to identify and isolate OCT images containing the fovea by outlining the retina tissue surface, applying filters to enhance image clarity, and using 2D and 3D selection modules to prioritize images with high fovea likelihood.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all OCT images are reviewed for diagnosis, then comprehensive retinal assessment is achieved, but diagnostic time and workload increase significantly
Solution Approach 1:
The system extracts and isolates only the most clinically relevant OCT images containing the fovea or near-fovea regions from the complete set of retinal images. By automatically identifying and separating these critical images based on anatomical landmarks and derivative calculations, the system enables clinicians to focus on the most diagnostically important slices without reviewing every image, thus reducing diagnostic time while maintaining assessment quality
Solution Approach 2:
The system applies different processing and selection criteria to different regions of the retina, with enhanced focus on the foveal region which is most critical for clinical diagnosis. By calculating derivatives and analyzing local anatomical features specifically in the foveal area, the system prioritizes images with high clinical relevance, allowing differentiated quality assessment across different retinal regions
2Productivity
If automated selection algorithms are implemented to identify fovea-containing images, then diagnostic efficiency improves, but system complexity increases
Solution Approach 1:
The system replaces manual visual inspection and subjective clinician judgment with automated computational algorithms that calculate derivatives of retinal tissue outlines and apply objective mathematical criteria for fovea identification. This substitution of mechanical/manual processes with automated computational methods improves diagnostic efficiency while the modular algorithm design keeps system complexity manageable
Solution Approach 2:
The system performs self-selection of relevant images by automatically analyzing its own input data through derivative calculations and probabilistic algorithms. The automated identification process uses intrinsic image features and mathematical transformations to autonomously determine which images contain the fovea, eliminating the need for external manual annotation or complex external processing systems
3Measurement precision
If high-resolution OCT imaging is used to capture detailed retinal structures, then image quality and diagnostic accuracy improve, but data volume and processing requirements increase
Solution Approach 1:
The system extracts only the essential diagnostic information by identifying and selecting a small subset of high-resolution images that contain the fovea, rather than processing and storing all high-resolution images. This extraction approach maintains the high measurement precision needed for accurate retinal assessment while significantly reducing the quantity of data that requires further processing and storage
Data Source
AI summary
The present disclosure describes systems and methods to select fovea containing optical coherence tomography (OCT) images. The systems and methods described herein receive a plurality of OCT images. The portion of the OCT images are selected for further processing, where a line tracing the border between the retina and non-retina tissue is generated. A difference of the line is generated. Candidate OCT images are then generated responsive to the generated difference line. The lowest point among each difference lines generated for each of the OCT images is identified, and the OCT image to which the lowest point corresponds is identified as the fovea containing OCT image.