Lung Region Division via Bronchial Branch Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In advanced lung diseases, the borders of lung sections become unclear, making it difficult to divide the lung region accurately using existing techniques, and changes in region size due to disease expansion or contraction obscure the recognition of diseased regions.
Innovation Solution
A region dividing device that extracts bronchi from a lung region and identifies branch positions to divide the lung into anatomical sections, such as upper, middle, and lower regions, and further divides into central, outer, inner, dorsal, and ventral regions based on bronchial branch positions, allowing independent division from predetermined percentages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If lung sections are divided using predetermined percentages based on distance from surface, then division is simple and automated, but accuracy deteriorates in advanced diseases where lung section borders become unclear
Solution Approach 1:
The patent introduces bronchial structures as an intermediary reference system to mediate the division process. Instead of directly dividing based on surface distance percentages, the system uses extracted bronchial trees as a mediating framework to define division planes, thereby maintaining automation while improving accuracy in diseased lungs where traditional anatomical landmarks are unclear
Solution Approach 2:
The patent replaces the mechanical/geometric division method (based on surface distance calculations) with a biologically-informed method using bronchial structure extraction. The division is no longer based on pure geometric percentages but on the actual bronchial branching patterns, substituting a mechanical system with one that adapts to biological variability
2Adaptability or versatility
If lung region size changes due to disease expansion or contraction, then the lung adapts to disease progression, but recognition of diseased regions deteriorates because size changes obscure lesion features
Solution Approach 1:
The patent segments the lung into multiple standardized regions based on bronchial branch positions. By dividing the lung into discrete segments (e.g., upper, middle, lower regions) defined by bronchial anatomy rather than overall lung size, the system can identify and analyze diseased regions within each segment independently, preventing size changes from obscuring lesion features
Solution Approach 2:
The patent applies local quality by analyzing each divided region separately with its own feature extraction and disease recognition parameters. Instead of treating the entire lung as a uniform region whose changing size affects overall recognition, each local segment is evaluated independently, allowing diseased regions to be recognized regardless of global lung size changes
3Measurement precision
If bronchial extraction and branch position identification is used for division, then division accuracy is improved in diseased lungs, but device complexity increases
Solution Approach 1:
The patent performs preliminary bronchial extraction and branch position identification before the actual division process. By pre-processing the image to extract bronchial structures and identify key branch positions in advance, the system prepares the necessary reference framework once, which then simplifies the subsequent division operations into standardized steps based on pre-identified anatomical landmarks
Data Source
AI summary
A lung region can be appropriately divided with a region dividing device, method, and program for dividing a lung region included in a medical image. A bronchus extracting unit 30 of a region dividing unit 22 extracts a bronchus from a lung region included in a medical image acquired by imaging a subject including a lung. A first dividing unit 31 to a fourth dividing unit 34 identify a plurality of branch positions of the bronchus and divide, based on the branch positions, the lung region into a plurality of regions.