Retinal Tomogram Artifact Correction for Layer Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing ophthalmic imaging technologies face challenges in accurately measuring retinal layer thickness and geometry due to artifacts caused by blood vessels and morbid tissues, which lead to difficulties in detecting layer positions and calculating layer thickness in regions with intensity attenuation.

Innovation Solution

An image processing apparatus and method that determines artifact regions by analyzing intensity statistical amounts and applies image correction to facilitate layer detection, using techniques such as edge detection, deformable models, and graph-based methods to accurately extract layer boundaries and correct intensity attenuation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image correction is applied to artifact regions, then layer detection accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvelayer detection accuracyVSAvoidimage processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The image processing apparatus segments the tomogram into artifact regions and non-artifact regions based on intensity statistical amounts. By dividing the processing area into distinct zones, the system applies different processing strategies to each segment, improving layer detection accuracy in artifact regions while managing overall processing complexity through localized treatment rather than global processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system applies image correction selectively only to artifact regions where intensity attenuation occurs, rather than processing the entire image uniformly. This local quality approach concentrates computational resources on problematic areas, enhancing layer detection accuracy where needed while avoiding unnecessary processing in unaffected regions, thus balancing accuracy improvement with complexity management.

Inventive Principle:
Principle #3Local quality

2Reliability

If artifact regions are identified and corrected, then measurement reliability is improved, but processing time increases

Engineering Contradiction:
Improvemeasurement reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary identification of artifact regions by analyzing intensity statistical amounts before conducting detailed layer detection. This preliminary action allows the system to pre-mark problematic areas and apply targeted correction only where necessary, improving measurement reliability while minimizing additional processing time by avoiding full-image reprocessing.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies image correction partially only to identified artifact regions rather than processing the entire image. This partial action approach focuses computational effort on the specific areas where attenuation occurs, improving measurement reliability in those regions while keeping the overall processing time increase minimal by leaving non-artifact regions uncorrected.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2437648B1Image processing apparatus, control method thereof, and computer program
Publication Date: 2019.09.11 CANON KK
  • EP2437648B1 patent drawingFigure 1A~1B
  • EP2437648B1 patent drawingFigure 2~3
  • EP2437648B1 patent drawingFigure 4

AI summary

The present invention an image processing apparatus, which processes an image of a tomogram obtained by capturing an image of an eye to be examined by a tomography apparatus, comprises, layer candidate detection means for detecting layer candidates of a retina of the eye to be examined from the tomogram, artifact region determination means for determining an artifact region in the tomogram based on image features obtained using the layer candidates, and image correction means for correcting intensities in the artifact region based on a determination result of the artifact region determination means and image features in the region.