Centerline Detection in 3D Medical Images Using Polar Coordinate Conversion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Creating a training set for artificial intelligence software to determine centerlines of elongated structures in three-dimensional medical images is a laborious and time-consuming process, requiring manual marking of centerlines in multiple slices, which hampers the efficiency of using AI for detecting abnormalities.

Innovation Solution

A system and method that generate training examples by marking reference points in a subset of slices, using an electronic processor to determine the centerline, convert cross-sections to polar coordinates, fit a line, and reconvert back to Cartesian coordinates, allowing for the creation of a training set that can automatically determine centerlines in elongated structures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual marking of centerlines in multiple slices is performed to create a training set, then the accuracy of training data is improved, but the time and labor required increases significantly

Engineering Contradiction:

Improveaccuracy of training dataVSAvoidtime to create training set

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides the image processing task into multiple slices or sections. Instead of manually marking centerlines across all slices, the system processes each slice independently and uses interpolation to estimate centerlines in intermediate slices. This segmentation allows automated processing of most slices while maintaining overall accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary automated centerline detection on a subset of slices before final training set creation. By pre-processing images to identify potential centerline locations and using these as reference points, the system reduces the manual marking burden while ensuring accurate training data generation.

Inventive Principle:

Principle #10Preliminary action

2Loss of information

If manual marking of centerlines in every slice is performed, then the completeness of training information is improved, but the labor intensity and complexity increase

Engineering Contradiction:

Improvecompleteness of training informationVSAvoidcomplexity of training set creation

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent enables the system to generate its own training data automatically. By using automated algorithms to detect centerlines in most slices and only requiring minimal manual input for validation, the system serves itself in creating comprehensive training sets without extensive human intervention at each step.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent introduces an automated centerline detection algorithm as an intermediary between manual marking and final training set creation. This intermediary process generates preliminary centerline estimates that are then refined through minimal manual correction, reducing both labor intensity and overall system complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If reference points are marked in every slice, then the precision of centerline determination is improved, but the productivity of training set creation decreases

Engineering Contradiction:

Improveprecision of centerline determinationVSAvoidspeed of training set creation

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent applies partial action by marking reference points in only a subset of slices rather than every slice. The automated interpolation algorithm then extends these partial markings to generate complete centerlines across all slices, maintaining sufficient precision while dramatically improving productivity.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The patent replaces the mechanical process of manual centerline marking in every slice with an automated computational system. The algorithm uses mathematical interpolation and image processing techniques to determine centerlines automatically, substituting manual mechanical marking with automated digital processing that is both faster and sufficiently accurate.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10916027B2Determining centerlines in elongated structures in images to detect abnormalities

Publication Date: 2021.02.09 MERATIVE US LP

AI summary

Systems and methods for determining an abnormality in an elongated structure in a three dimensional medical image. One system includes an electronic processor. The electronic processor is configured to determine a centerline of the elongated structure in the three dimensional medical image and determine a plurality of two dimensional cross sections of the three dimensional medical image based on the centerline. For each two dimensional cross section of the plurality of two dimensional cross sections, the electronic processor is configured to convert the two dimensional cross section to polar coordinates, fit a line to the elongated structure in the two dimensional cross section converted to polar coordinates, and reconvert the two dimensional cross section to Cartesian coordinates.