Endoscope Lesion Assessment via Color Space Angle Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing endoscope systems face challenges in objectively quantifying the extent of lesions in biological tissue due to fluctuations in inflammation evaluation values caused by image brightness and difficulty in distinguishing changes in lesion extent along the depth direction of an organ.
Innovation Solution
An electronic endoscope system with an evaluation value calculation unit, assessment unit, and representative value determination unit processes images to assess and quantify the extent of lesions by analyzing variations in evaluation values and using indicators such as difference, standard deviation, or regression line fit to determine if the lesion extent has changed, and defines representative values for each section.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If inflammation evaluation values are calculated based on image brightness, then the evaluation can be performed automatically, but the evaluation values fluctuate due to brightness changes and cannot accurately indicate lesion extent
Solution Approach 1:
The patent changes the parameter used for evaluation from brightness-based metrics to color-based metrics (specifically the angle in color space). This allows automatic evaluation while eliminating sensitivity to brightness changes, as color ratios remain consistent regardless of illumination intensity.
Solution Approach 2:
The patent substitutes the mechanical/optical measurement approach (brightness-based evaluation) with a color space analysis approach. By transforming the evaluation from luminance-based to chromaticity-based, the system achieves automatic processing without brightness sensitivity.
2Loss of information
If evaluation values are calculated for each image, then detailed lesion information is obtained, but it is difficult to determine whether variations are due to brightness or actual lesion changes
Solution Approach 1:
The patent changes from using absolute brightness values to using color space angles (chromaticity coordinates). This parameter transformation preserves lesion information while eliminating the confounding effect of brightness variations, making the evaluation reliable for detecting actual lesion changes.
3Measurement precision
If color-based evaluation is performed to distinguish lesion parts, then objective quantification is achieved, but long-term training is required for surgeons to interpret color differences
Solution Approach 1:
The patent utilizes color changes in the biological tissue as the basis for automated evaluation. By calculating the angle in color space relative to a reference axis, the system automatically quantifies lesion extent without requiring surgeons to interpret subtle color differences, thus eliminating the need for extensive training.
Solution Approach 2:
The patent substitutes human visual interpretation of color differences with automated color space analysis. The calculation of angles in color space provides objective quantification that eliminates the need for surgeon training in color interpretation while maintaining high detection accuracy.
4Loss of information
If multiple images are analyzed to assess lesion extent along depth direction, then comprehensive evaluation is achieved, but processing load increases
Solution Approach 1:
The patent changes from analyzing multiple full-image datasets to analyzing extracted color space parameters (angles) from each image. This parameter reduction allows comprehensive evaluation of lesion extent across multiple images while significantly reducing processing load, as angle calculations are computationally efficient.
Data Source
AI summary
This electronic endoscope system includes an evaluation value calculation unit, an assessment unit, and a representative value determination unit. The evaluation value calculation unit is configured so as to obtain an evaluation value indicating an extent of a lesion in biological tissue in each of a plurality of images captured in a predetermined section along the depth direction of a region in an organ. The assessment unit is configured so as to assess whether the extent of the lesion in the section is changed on the basis of the degree of variation of the evaluation value. The representative value determination unit is configured so as to define a representative value of the section representing the evaluation value, in a different method when the extent of the lesion is assessed to be changed and when the extent of the lesion is assessed not to be changed.