Endoscope LED Light Amount Adjustment for Lesion Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional endoscope devices cannot improve determination accuracy by adjusting the light amount for regions of interest suspected to be lesions, as identified by learning models.

Innovation Solution

A computer program that acquires images from an endoscope, inputs them into a learning model to recognize regions of interest, and adjusts the light emitted by LEDs at the endoscope's distal tip based on the recognition results, including adjusting the light amount and displaying the region of interest separately when the likelihood of a lesion is low or high.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the light amount for a region of interest is adjusted based on learning model recognition, then the determination accuracy of lesions is improved, but the device complexity increases

Engineering Contradiction:
Improvedetermination accuracyVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system uses the learning model to automatically identify regions of interest and determine appropriate light amounts without manual intervention. The control unit autonomously processes images, recognizes lesions, and adjusts illumination based on pre-established criteria, enabling the system to serve itself in the diagnostic process.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system dynamically changes the light amount parameter based on the recognition results from the learning model. When a region of interest is identified, the control unit adjusts the light amount to a predetermined value, transforming the static illumination system into a dynamic one that adapts to detected conditions.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If multiple LEDs are used to irradiate different regions, then the adaptability of illumination adjustment is improved, but the device complexity increases

Engineering Contradiction:
Improveadaptability of illumination adjustmentVSAvoiddevice complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The illumination system is divided into multiple independent LEDs, each capable of emitting light to specific regions. This segmentation allows independent control of light amount for different areas, enabling precise and adaptable illumination adjustment based on the location and characteristics of detected regions of interest.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different LEDs are assigned to irradiate different regions with specific light amounts based on local requirements. The control unit determines which LED(s) to activate and at what intensity, providing localized illumination quality tailored to the specific needs of each region of interest rather than uniform illumination throughout.

Inventive Principle:
Principle #3Local quality

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

This solution allows for automatic adjustment of light amounts to improve the accuracy of lesion detection by optimizing illumination for regions of interest, enhancing the determination of lesions.

Implementation Method 1

adjusting a light amount with which an LED provided at a distal tip of the endoscope irradiates the region of interest

Methodology Applied
Scientific EffectLight Emitting Diode: Light Emitting Diode

Data Source

PatentUS20240415379A1Computer program, information processing method, and endoscope
Publication Date: 2024.12.19 HOYA CORPORATION
  • US20240415379A1 patent drawing
  • US20240415379A1 patent drawing
  • US20240415379A1 patent drawing

AI summary

A computer program causes a computer to execute processing including: acquiring an image captured by an endoscope; inputting, when the image captured by the endoscope is input, the image to a learning model learned to output a recognition result of a region of interest included in the image, and outputting a recognition result; and adjusting a light amount with which an LED provided at a distal tip of the endoscope irradiates the region of interest, on the basis of the recognition result.