Hyperdense Lung Tissue Detection Using Vessel Neighbour Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated detection of hyperdense regions in lung images is challenging due to atypical densities caused by diseases, leading to poor lung segmentation and difficulty in distinguishing between hyperdense parenchyma and pleural effusions.
Innovation Solution
A system that locates vessels in the image and focuses on regions neighboring the vessels to identify hyperdense lung tissue, using density analysis to determine if the regions are hyperdense parenchyma or pleural effusions, thereby improving computational efficiency and reducing false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If standard lung segmentation algorithms are used, then automated detection can be performed, but hyperdense regions are not recognised due to atypical high densities leading to poor segmentation
Solution Approach 1:
The patent divides the lung image into multiple density-based regions (aerated lung tissue, hyperdense regions, pleural effusions) and processes each region separately with appropriate algorithms, allowing accurate identification of hyperdense regions that standard segmentation would miss
Solution Approach 2:
The patent changes the density threshold parameters used in segmentation to specifically identify hyperdense regions (>0 HU) separately from normal aerated lung tissue, enabling the algorithm to recognize and properly segment atypical high-density areas
2Productivity
If density-based detection is used to identify hyperdense regions, then automated detection is achieved, but it becomes challenging to distinguish betweenhyperdense parenchyma and pleural effusions
Solution Approach 1:
The patent introduces an intermediary analysis step that examines the relationship between hyperdense regions and vascular structures. By checking whether hyperdense regions contain or neighbour vessels, the system distinguishes hyperdense parenchyma (which contains vessels) from pleural effusions (which do not), resolving the classification ambiguity
Solution Approach 2:
The patent applies different analysis criteria to different spatial relationships: hyperdense regions containing vessels are classified as hyperdense parenchyma, while those without vessels are classified as pleural effusions, allowing accurate local classification based on vascular presence
3Measurement precision
If manual sampling of density values is used, then accurate identification can be achieved, but it is time consuming and suffers from reproducibility issues
Solution Approach 1:
The patent implements an automated system that performs density sampling, region identification, and classification without manual intervention. The algorithm automatically samples density values, identifies hyperdense regions, distinguishes parenchyma from effusions using vascular markers, and produces results with both high accuracy and reproducibility
Solution Approach 2:
The patent replaces the manual mechanical sampling process with an automated computational algorithm that systematically analyzes density values across the entire lung image, eliminating time-consuming manual operations while maintaining and improving measurement accuracy
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided a computer-implemented method and system (100) for determining regions of hyperdense lung parenchyma in an image of a lung. The system (100) comprises a memory (106) comprising instruction data representing a set of instructions and a processor (102) configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor (102), cause the processor (102) to locate a vessel in the image, determine a density of lung parenchyma in a region of the image that neighbours the located vessel, and determine whether the region of the image comprises hyperdense lung parenchyma based on the determined density, hyperdense lung parenchyma having a density greater than -800 HU.