Graphical Segmentation Interface Optimizing User Interactions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for image segmentation, such as those described in US 2015/0155010 A1, are insufficiently efficient for users, particularly in medical imaging, as they rely on video-based instructions and do not optimize user interactions for faster and more accurate segmentation.

Innovation Solution

A system and method that analyze user interaction data to determine an optimized set of interactions using a graphical segmentation interface, generating candidate sets based on similarity and time metrics to provide users with a faster and more efficient segmentation process, potentially introducing new tools or parameter settings not initially used.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If video-based instructions are provided to help users operate segmentation tools, then user guidance is improved, but user efficiency in obtaining segmentation remains insufficient

Engineering Contradiction:
Improveuser guidanceVSAvoiduser efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The system records user interactions with segmentation tools and provides feedback by suggesting optimized interaction sequences that achieved similar segmentations faster, allowing users to learn from actual optimized workflows rather than passive video instructions

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system automatically analyzes user interaction data and generates optimized interaction suggestions without requiring external training materials, enabling the system to self-improve and provide personalized efficiency enhancements to each user

Inventive Principle:
Principle #25Self-service

2Measurement precision

If manual or semi-automatic interaction is used to correct automated segmentation, then segmentation accuracy is improved, but time consumption increases

Engineering Contradiction:
Improvesegmentation accuracyVSAvoidtime consumption
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system pre-analyzes user interaction patterns and prepares optimized interaction sequences in advance, so when users need to correct automated segmentation, they can immediately apply proven efficient interaction patterns rather than experimenting in real-time

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system varies interaction parameters such as tool selection, parameter settings, and interaction sequences to find optimal combinations that achieve accurate segmentation faster, adapting to different segmentation scenarios and user skills

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If multiple segmentation tools with varying parameters are provided, then segmentation flexibility is improved, but interaction complexity increases

Engineering Contradiction:
Improvesegmentation flexibilityVSAvoidinteraction complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system makes segmentation tools multi-functional by enabling each tool to operate with optimized parameters for different scenarios, reducing the need for users to manually configure multiple specialized tools while maintaining segmentation flexibility

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system automatically adjusts tool parameters based on the segmentation context and user interaction patterns, allowing a smaller set of tools to achieve the same versatility as many specialized tools would require, thereby reducing interaction complexity

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3378037B1Optimizing user interactions in segmentation
Publication Date: 2023.07.26 KONINKLIJKE PHILIPS NV
  • EP3378037B1 patent drawingFigure 1
  • EP3378037B1 patent drawingFigure 2a~2c
  • EP3378037B1 patent drawingFigure 3

AI summary

A system and a computer-implemented method are provided for segmenting an object in a medical image using a graphical segmentation interface. The graphical segmentation interface may comprise a set of segmentation tools for enabling a user to obtain a first segmentation of the object in the image. This first segmentation may be represented by segmentation data. Interaction data may be obtained which is indicative of a set of user interactions of the user with the graphical segmentation interface by which the first segmentation of the object was obtained. The system may comprise a processor configured for analyzing the segmentation data and the interaction data to determine an optimized set of user interactions which, when carried out by the user, obtains a second segmentation similar to the first segmentation, yet in a quicker and more convenient manner. A video may be generated for training the user by indicating the optimized set of user interactions to the user.