Atrial Wall Thickness Mapping from CT Mask Labels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing atrial wall thickness measurement methods are not applicable to commercially available automatic segmentation software and require manual clinician input, leading to inconsistent results due to dependence on clinician experience.
Innovation Solution
A method and device using a processor and memory to calculate atrial wall thickness through image data analysis, including steps like convex hull algorithm, Poisson equation, and Laplace equation, to automatically define endocardial and epicardial borders without manual input, utilizing mask label image data from CT scans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If manual clinician input is used to define endocardial and epicardial borders, then measurement can be performed, but consistency and reliability of results deteriorate due to dependence on clinician experience
Solution Approach 1:
The system performs self-service by automatically defining endocardial and epicardial borders using image processing algorithms without requiring manual clinician input. The processor autonomously segments the atrial wall and calculates thickness measurements, eliminating the need for experienced clinicians to manually specify borders while maintaining measurement consistency across different users.
Solution Approach 2:
The patent replaces the mechanical/manual system of clinician-based border definition with an automated image processing system. Algorithms including convex hull, Poisson equation, and Laplace equation substitute for manual clinician operations, transforming the measurement process from a skill-dependent manual task to an automated computational process that delivers consistent results.
2Adaptability or versatility
If customized atrial wall segmentation technology is used, then atrial wall thickness can be measured, but adaptability deteriorates because it is not applicable to commercially available automatic segmentation software
Solution Approach 1:
The patent achieves universality by designing an automated border definition system that can operate with multiple types of input data from different commercial segmentation software packages. The system accepts mask label image data and other standard formats, making it compatible with various commercial tools while maintaining its automated measurement capabilities, thus eliminating the need for customized segmentation technology.
3Productivity
If manual specification of endocardium and epicardium is required, then measurement can be performed, but productivity deteriorates due to time-consuming manual operations
Solution Approach 1:
The system performs preliminary action by automatically completing the border definition process before the clinician needs to review or approve measurements. The processor pre-calculates endocardial and epicardial borders using image processing algorithms, eliminating the need for time-consuming manual border specification and significantly accelerating the overall measurement workflow.
Data Source
AI summary
A method of calculating an atrial wall thickness through a device including a processor and a memory according to an embodiment of the present disclosure may include (a) receiving image data for a patient's atrium as input data, (b) calculating endocardial and epicardial borders corresponding to starting and ending points of the atrial wall thickness from the received input data, and (c) calculating the patient's atrial wall thickness using the calculated endocardial and epicardial borders, wherein the image data for the patient's atrium received in the step (a) is a plurality of mask label image data for respective tomographic images of the patient's atrium acquired from computed tomography (CT) scans for the patient's atrium, and the mask label image data is data in which one or more pixels corresponding to an atrial wall region including the patient's atrium and an inner region of the atrium, other regions within the patient's heart disposed around the patient's atrium, and regions other than the patient's heart, respectively, are assigned with labels having different numerals for the respective regions.