Endoscopic Image Analysis Device for Polyp Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current medical image analysis systems, particularly endoscopic devices, rely heavily on subjective interpretation and struggle to accurately detect polyps due to limitations in edge detection methods, which fail when observing polyps from certain angles and can mistakenly identify lumen folds as polyps.

Innovation Solution

An image analysis device that processes endoscopic images by calculating gray value gradients and isotropic change feature values to detect polyps without relying on edge detection, using a CPU to generate and display polyp images based on these calculations, allowing for two-dimensional analysis and accurate polyp localization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If edge detection methods are used to detect polyps, then the detection process is simple, but the detection accuracy deteriorates when observing polyps from certain angles and lumen folds are mistakenly identified as polyps

Engineering Contradiction:

Improvedetection process simplicityVSAvoidpolyp detection accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent changes the detection parameters from edge-based features to gray value gradient-based features. By calculating gray value gradients in multiple directions (horizontal, vertical, diagonal) and comparing their relationships, the system achieves more reliable polyp detection that is not affected by viewing angle or mistaken identification of lumen folds.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If subjective interpretation by doctors is used for diagnosis, then the diagnostic process is flexible, but the diagnosis reliability deteriorates due to subjectivity

Engineering Contradiction:

Improvediagnostic process flexibilityVSAvoiddiagnosis reliability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent replaces the mechanical/subjective interpretation process with an automated computational system. The CPU automatically calculates gray value gradients, compares directional relationships, and determines polyp presence based on objective numerical criteria, eliminating doctor subjectivity while maintaining diagnostic flexibility through programmable analysis methods.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If conventional image analysis methods are used, then the analysis speed is fast, but the lesion determination accuracy deteriorates

Engineering Contradiction:

Improveanalysis speedVSAvoidlesion determination accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent segments the image analysis process into distinct computational stages: calculating horizontal and vertical gray value gradients, calculating diagonal gradients, comparing directional relationships, and making determination decisions. This segmented approach maintains processing speed while improving accuracy through systematic multi-directional analysis.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS8244009B2Image analysis device

Publication Date: 2012.08.14 OLYMPUS CORPORATION(JP)

AI summary

A CPU executes a gray value gradient calculation process to, for example, an R signal in step S10, executes an isotropic change feature value calculation process based on a gray value gradient calculated in step S11, and further executes a possible polyp detection process for generating a possible polyp image at a location where a polyp exists based on an isotropic change feature value calculated in step S12. This improves the detection of an existing location of an intraluminal abnormal tissue.