3D Medical Image Segmentation Seed Point Refinement

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for determining a volume of interest (VOI) in 3D medical images are inconsistent and inaccurate due to user-dependent seed point selection, leading to variability in segmentation results.

Innovation Solution

A system and method for image segmentation that acquires a 3D image, determines a preliminary seed point, and iteratively refines it to a final seed point based on geometric or gravitational characteristics, using a combination of pixel gray value analysis and user input to stabilize the segmentation process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If user manually selects seed point based on experience, then segmentation can be initiated, but determination of VOI becomes inconsistent and inaccurate

Engineering Contradiction:

Improveaccuracy of VOI determinationVSAvoidconsistency of segmentation results

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The system enables automatic seed point determination by having the algorithm select its own seed point based on image characteristics (such as maximum gray value pixels) rather than relying on user selection. This self-service approach eliminates human variability and improves both accuracy and consistency of VOI determination across different users and sessions.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent changes the parameter selection criterion from user-defined spatial coordinates to image-based parameters such as pixel gray values. By selecting the seed point at the location of maximum gray value or according to specific intensity thresholds, the system transforms the seed point selection from a subjective manual process to an objective automated process based on measurable image parameters.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If automatic seed point determination is implemented, then consistency improves, but user control and adaptability may be reduced

Engineering Contradiction:

Improveconsistency of segmentation resultsVSAvoiduser control over seed point selection

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system provides dynamic adaptability by allowing users to adjust segmentation parameters and re-initiate the automatic seed point determination process as needed. The algorithm can be re-executed with modified parameters to adapt to different imaging conditions or clinical requirements, maintaining both automation benefits and user control flexibility.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms where users can review the automatically determined seed point and VOI results, then provide corrections or adjustments. This feedback loop allows the system to learn from user inputs and refine its automatic determination algorithm, balancing automation with user expertise.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS10417767B2Systems and methods for image segmentation

Publication Date: 2019.09.17 SHANGHAI UNITED IMAGING HEALTHCARE

AI summary

A method for image segmentation includes acquiring a three-dimensional (3D) image that includes a plurality of two-dimensional (2D) images arranged in a spatial order. The method also includes determining a preliminary seed point in a first 2D image of the plurality of 2D images. The method further includes determining, based on the preliminary seed point, a final seed point in a second 2D image of the plurality of 2D images, and determining, based on the final seed point, a volume of interest (VOI) in the 3D image.