Chest Motion Analysis for COPD Severity Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing Chronic Obstructive Pulmonary Disease (COPD) using chest CT images are limited in accuracy, as they rely on CT values that are not consistent across all CT machines and struggle to accurately determine disease severity from voxel motion and CT value changes between expiration and inspiration images.
Innovation Solution
A medical image processing apparatus that calculates the motion amount of chest area pixels or regions between different respiratory phases, combining this data with feature values and area size change rates to determine COPD severity levels and output relevant information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If determination is performed based on a given threshold at a given CT value, then the determination process is simple, but the diagnosis accuracy deteriorates because CT values are not consistent across all CT apparatuses
Solution Approach 1:
The patent transforms the determination criteria from fixed CT value thresholds to motion-based parameters. Instead of using absolute CT values that vary between apparatuses, the system calculates motion amounts (displacement distances) of voxels between inspiration and expiration phases, and uses these motion parameters for determination. This parameter transformation eliminates apparatus-specific calibration issues while maintaining determination simplicity.
2Device complexity
If determination is performed using only motion amounts or CT values from expiration and inspiration images, then the analysis process is simple, but the COPD diagnosis accuracy deteriorates
Solution Approach 1:
The patent merges multiple parameters for COPD determination: motion amounts (displacement distances) of voxels, CT values at different phases, and the combination of these parameters. By integrating both motion-based and CT value-based information, the system achieves more accurate local COPD diagnosis while maintaining a manageable analysis process that can be implemented through systematic calculation steps.
3Loss of information
If local COPD diagnosis is performed for each pixel or lung area, then the diagnostic information becomes more useful for treatment planning, but the processing complexity increases compared to overall lung field diagnosis
Solution Approach 1:
The patent segments the lung field into multiple regions (lung lobes, pulmonary segments, or voxel-level divisions) and performs motion amount calculations and COPD determination for each segment independently. This segmentation approach provides detailed local diagnostic information useful for treatment planning while organizing the processing into manageable segments that reduce overall system complexity.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
Support information effective in diagnosing chronic obstructive pulmonary disease is provided from a chest volume image. A storage device (112) of a medical image processing apparatus stores the data of a plurality of images in different respiratory phases which represent the morphology of an object chest portion. A vector calculation processing unit (121) calculates the motion amount of a region between the plurality of images for each pixel or area. A level decision unit (123) decides a level concerning the severity of chronic obstructive pulmonary disease for each pixel or area based on a combination of at least two of a motion amount, a feature value obtained from pixel values of image, and the change rate of the size of the area.