Fluorescence Endoscope Image Correction via Reference Division
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fluorescence endoscope devices face challenges in accurately distinguishing between diseased and healthy tissue due to variations in fluorescence intensity caused by observation distance and angle, leading to false positives and background noise in images.
Innovation Solution
A fluorescence endoscope device that captures both fluorescence and reference images, divides the fluorescence image by the reference image to correct for intensity variations, and sets thresholds based on pixel gradation values and standard deviations to extract and correct regions, thereby removing background noise and false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fluorescence image is captured to detect diseased sites, then disease detection capability is improved, but false positives and background noise increase due to intensity variations from observation distance and angle
Solution Approach 1:
The patent introduces a reference image as an intermediary element to mediate between the fluorescence image and the true disease state. The reference image, captured under illumination light without fluorochrome excitation, serves as a mediator to identify and remove background noise and false positives caused by observation distance and angle variations, thereby improving both measurement precision and reliability
Solution Approach 2:
The patent changes the parameter of light type used for image capture by capturing both fluorescence images (using excitation light) and reference images (using illumination light). This parameter change enables comparison and differentiation between true fluorescence signals and background noise, resolving the contradiction between detection accuracy and image reliability
2Measurement precision
If fluorescence intensity is used to identify diseased regions, then disease localization is improved, but background regions and false positives are incorrectly identified
Solution Approach 1:
The reference image acts as an intermediary to distinguish true disease signals from false positives. By comparing fluorescence image intensity with reference image intensity at corresponding pixels, the system can identify regions where high fluorescence intensity is due to background noise rather than actual disease, thereby reducing false positives while maintaining accurate disease localization
Solution Approach 2:
The patent implements a feedback mechanism where the reference image provides information about background intensity levels that feeds back into the interpretation of fluorescence image data. This feedback loop enables the system to adjust its disease identification criteria dynamically, reducing false positives while maintaining sensitivity to true disease signals
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
The device generates a corrected fluorescence image that effectively isolates diseased sites by suppressing background and false positive regions, improving image quality and accuracy by reducing the influence of observation distance and angle variations.
Implementation Method 1
irradiating an observation target site doped with a fluorochrome that preferentially accumulates in a diseased site, such as a cancer cell, with excitation light for exciting the fluorochrome to generate drug fluorescence
Implementation Method 2
captures an image of return light returning from the subject due to the irradiation with the illumination light from the light source
Data Source
AI summary
Provided is a fluorescence endoscope device that includes a light source; an image generating portion that captures an image of fluorescence generated at a subject due to irradiation with excitation light to obtain a fluorescence image and that captures an image of return light returning from the subject due to irradiation with white light to obtain a white-light image; a dividing portion that divides the fluorescence image by the white-light image to generate a divided fluorescence image; a coordinate extracting portion that extracts a second region of the divided fluorescence image having a gradation value higher than a second threshold; a fluorescence-image correcting portion that extracts a first region having a gradation value higher than a first threshold in the fluorescence image and generates a corrected fluorescence image in which an overlap region that overlaps the second region is extracted; and a monitor that displays the corrected fluorescence image.