Medical Test Image Analysis with AI Test Area Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing medical image analysis methods, particularly for ECG and chest X-ray images, lack specialized techniques for accurate and efficient selection and analysis, hindering high user convenience and effective diagnosis.

Innovation Solution

An apparatus and method utilizing an artificial neural network-based detection model to automatically select and analyze test areas in ECG and chest X-ray images, providing natural language descriptions and digital biomarker results, with features like confidence scoring and user-driven boundary box adjustments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If deep learning-based image analysis models are used, then analysis accuracy is improved, but device complexity increases

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

Solution Approach 1:

The system segments the medical image analysis task into distinct functional modules: a detection model for identifying test areas and generating boundary boxes, and a separate analysis model for interpreting the selected regions. This modular segmentation allows each component to be optimized independently, maintaining high accuracy while managing system complexity through organized functional decomposition.

Inventive Principle:
Principle #1Segmentation

2Productivity

If automated test area selection is implemented, then productivity is improved, but measurement precision may worsen

Engineering Contradiction:
Improveanalysis efficiencyVSAvoidtest area selection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The detection model incorporates confidence score feedback to evaluate the quality of generated test area selections. By automatically assessing the reliability of each detected test area through confidence metrics, the system can prioritize high-quality selections for automated analysis while maintaining precision standards, thus achieving both efficiency and accuracy.

Inventive Principle:
Principle #23Feedback

3Adaptability or versatility

If multiple types of medical images are supported, then adaptability is improved, but device complexity increases

Engineering Contradiction:
Improveimage type supportVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system implements universal detection and analysis models that can process multiple types of medical images (ECG, chest X-ray, and other test result images) through a common framework. The detection model generates boundary boxes applicable to various image types, and the analysis model adapts to different modalities, achieving multi-functionality without proportionally increasing system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4708192A1Apparatus and method for analyzing medical examination result images
Publication Date: 2026.03.11 SEOUL NAT UNIV HOSPITAL
  • EP4708192A1 patent drawingFigure 1
  • EP4708192A1 patent drawingFigure 2A
  • EP4708192A1 patent drawingFigure 2B

AI summary

Disclosed are a medical examination result image analysis apparatus and a medical examination result image analysis method performed thereby. The method comprises the steps of: selecting, by a processor, by using an artificial neural network-based detection model, a test area to be provided for analysis from among medical test result images displayed on a screen; and analyzing, by the processor, the selected test area by using an artificial neural network-based analysis model, providing results of the analysis, wherein the medical test result images may include one or more of an electrocardiogram image and a chest X-ray image.