Endoscope Oxygen Saturation Certainty Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing endoscope systems face challenges in accurately determining the certainty of oxygen saturation in biological tissue, leading to potential misidentification of lesioned areas due to low certainty values, especially when pixel values are small or large, which can result in unreliable image display.
Innovation Solution
An endoscope system that uses a light source emitting at least two types of light with different wavelength bands to generate color image data, where a processor calculates oxygen saturation and certainty based on components of the image data, setting threshold values for certainty calculation to filter out unreliable data and display only high-certainty oxygen saturation distribution images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If oxygen saturation is calculated using image data processing, then oxygen saturation distribution images can be generated to help identify lesioned parts, but pixel values that are very small or very large exist in some cases leading to low certainty of the calculated oxygen saturation
Solution Approach 1:
The patent changes the parameters used for certainty calculation by introducing multiple image data components (first component, second component, third component) with different relationships to oxygen saturation. By calculating certainty based on multiple different parameter combinations rather than a single parameter, the system achieves more reliable certainty assessment across varying pixel value ranges.
Solution Approach 2:
The patent combines multiple image data components to form a composite certainty value. Instead of relying on a single parameter, the system integrates information from multiple components (including their ratios and differences) to create a composite measure of certainty, thereby improving overall reliability of the oxygen saturation measurement.
2Reliability
If information on the certainty of oxygen saturation is obtained based on image data, then the operator can judge with higher confidence, but the information on the certainty may become excessive if not efficiently obtained
Solution Approach 1:
The patent extracts only the essential components needed for certainty calculation from the image data. By selectively using specific components (first component, second component, third component) and their relationships rather than processing all possible image data, the system obtains sufficient certainty information without excessive processing complexity.
Solution Approach 2:
The patent applies partial action by calculating certainty using a specific subset of image data components rather than all available data. This selective approach provides adequate certainty information for clinical decision-making while avoiding the computational burden of processing excessive data.
3Area of stationary object
If all pixel data is displayed in the oxygen saturation distribution image, then complete coverage of the tissue is achieved, but low-certainty pixels may lead to misidentification of lesioned areas
Solution Approach 1:
The patent applies local quality by differentiating the display treatment of pixels based on their certainty values. High-certainty pixels are displayed with full visibility while low-certainty pixels are displayed with reduced visibility or excluded. This localized differentiation ensures that the displayed image maintains high reliability for lesion identification while still providing comprehensive spatial coverage.
Data Source
AI summary
A processor of an endoscope system uses components of color image data of biological tissue illuminated by at least two types of light to generate an oxygen saturation distribution image of hemoglobin in the biological tissue and to calculate a certainty of an oxygen saturation. Furthermore, the processor controls a display mode of the oxygen saturation distribution image according to the degree of the certainty. The certainty drops if the value of a first component a of the color image data is lower than a lower limit threshold value. The certainty drops if the value of a second component b of the color image data is greater than an upper limit threshold value. The lower limit threshold value and the upper limit threshold value are values of two different pieces of color image data.