Endoscope Image Processing for Atrophic Gastritis Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the diagnosis of atrophic gastritis, existing endoscope systems face challenges in clearly distinguishing between normal and atrophic gastric mucosa, especially in early stages where the color differences are subtle, leading to difficulties in determining the progression of atrophy and identifying lesions like gastric cancer.
Innovation Solution
An image processing device and method that enhance color differences and structures by calculating specific signal ratios between RGB image signals, generating frequency-enhanced images, and composing them with base images to highlight abnormal areas, such as faded mucosa, blood vessels, and reddened regions, using a combination of color difference and structure enhancement techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-vision system with increased resolution is used, then image quality and ability to depict fine structures is improved, but ability to detect subtle color differences in early lesions is not sufficiently improved
Solution Approach 1:
The patent applies color enhancement processing that selectively enhances the color of suspected lesion areas to make subtle color differences more visible. By detecting slight color variations and amplifying them through processing, the system enables doctors to distinguish early lesions that would be imperceptible in standard images, directly addressing the limitation of resolution-only approaches.
Solution Approach 2:
The system changes the parameter of color intensity in processed images by enhancing the color saturation and contrast of suspected lesion areas. This parameter transformation converts subtle, hard-to-detect color differences into pronounced visual differences, allowing detection of early gastric cancer lesions that maintain normal mucosal appearance at standard resolution.
2Difficulty of detecting and measuring
If color enhancement processing is applied to enhance lesion visibility, then detection of reddish lesion parts is improved, but natural color representation of the mucosa is altered
Solution Approach 1:
The system performs preliminary detection of suspected lesion areas based on initial color analysis before applying enhancement processing. By first identifying regions with abnormal color characteristics and then selectively enhancing only those specific areas, the system preserves the natural appearance of normal mucosa while highlighting potential lesions, thus maintaining reference to original color information.
Solution Approach 2:
The enhancement processing is applied locally only to suspected lesion areas rather than uniformly across the entire image. This localized approach enhances color differences in abnormal regions while leaving normal mucosa unchanged, allowing doctors to compare enhanced regions against unenhanced reference areas for accurate diagnosis.
3Difficulty of detecting and measuring
If frequency band-enhancement processing is performed according to discolored area size, then structure of discolored areas is enhanced, but processing complexity increases
Solution Approach 1:
The system applies frequency band enhancement selectively based on the detected size and characteristics of discolored areas. Rather than applying uniform enhancement across all images, the processing intensity and frequency bands are adjusted according to the specific characteristics of each suspected lesion, optimizing detection effectiveness while controlling processing complexity through adaptive parameter selection.
Data Source
AI summary
A base image is created from RGB image signals. A B/G ratio between the B image signal and the G image signal is calculated, and a G/R ratio between the G image signal and the R image signal is calculated. A frequency component-enhanced image in which a frequency component corresponding to an abnormal part different from a normal part is enhanced is generated based on the base image. A composition ratio which indicates the ratio for composing the frequency component-enhanced image with the base image is set based on the B/G ratio and the G/R ratio. The frequency component-enhanced image is composed with the base image at the set composition ratio to generate a structure-enhanced image in which the structure of the abnormal part is enhanced.