Pulmonary Emphysema Region Extraction Using Segmentation and Distribution Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing medical image diagnosing technologies for pulmonary emphysema lack accuracy in extracting the pulmonary emphysema region, often resulting in excessive pixel extraction and reduced precision.

Innovation Solution

A method involving multiple steps: site region extraction, first and second lesion candidate region extraction based on pixel values, and region correction using distribution analysis to improve the accuracy of pulmonary emphysema region detection and display.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If threshold-based extraction method is used to extract pulmonary emphysema region, then extraction speed is improved, but extraction precision deteriorates due to excessive pixel extraction

Engineering Contradiction:

Improveextraction speedVSAvoidextraction precision

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The extraction process is divided into multiple stages: initial threshold-based extraction to get candidate regions, followed by secondary refinement using distribution analysis and morphological operations. This segmentation allows fast initial extraction while maintaining precision through subsequent processing steps.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A candidate region extraction step serves as an intermediary between raw threshold extraction and final precise region identification. This intermediate step filters out excessive pixels by using distribution characteristics and morphological operations, bridging the gap between speed and precision.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If simple threshold extraction is used, then device complexity is reduced, but extraction accuracy deteriorates

Engineering Contradiction:

Improveprocessing complexityVSAvoidregion extraction accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The processing pipeline is segmented into distinct modules: threshold-based candidate extraction, distribution analysis, morphological operations, and boundary refinement. Each module performs a specific function with manageable complexity, while collectively achieving high extraction accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Threshold-based extraction is performed as a preliminary action to quickly identify candidate regions before applying more complex refinement operations. This preliminary step reduces the data volume for subsequent processing while maintaining the foundation for accurate extraction.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS7940975B2Medical image diagnosing support method and apparatus, and image processing program configured to extract and correct lesion candidate region

Publication Date: 2011.05.10 FUJIFILM CORP

AI summary

A medical image diagnosing support method is provided that includes a site region extracting step for obtaining a tomographic image which is picked up by a medical diagnostic imaging apparatus and extracting a predetermined site region from the obtained tomographic image, a first region extracting step for extracting a first lesion candidate region from the site region based on pixel values of the site region extracted in the predetermined site region extracting step, a second region extracting step for extracting a second lesion candidate region from the site region based on a distribution of the pixel values of the site region extracted in the predetermined site region extracting step, and a region correcting step for correcting the first lesion candidate region extracted in the first region extracting step by using the second lesion candidate region extracted in the second region extracting step.