Neutrosophic Image Segmentation for Surgical Registration

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional real-time visualization techniques in medical procedures face challenges in accurately importing prior analysis onto real-time images due to the static and less accurate nature of co-registered images, which can lead to inaccuracies during surgical resections.

Innovation Solution

A region growing algorithm based on neutrosophic similarity scores is used for image segmentation and registration, allowing for the importation of prior analysis onto real-time images by transforming pixel characteristics into neutrosophic set domains and calculating similarity scores to segment objects and register images accurately.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If co-registration techniques are used to import prior images, then prior analysis can be displayed, but the images become static and inaccurate due to tissue changes and position variations

Engineering Contradiction:

Improveprior analysis import capabilityVSAvoidimage accuracy

Core Design Contradiction:

Loss of informationVSReliability

Solution Approach 1:

The system transitions from static co-registered images to dynamic real-time image fusion. Real-time imaging modalities (ultrasound, MRI, CT) continuously update the visualization to reflect current tissue positions and states, eliminating the static nature of traditional co-registration while maintaining the ability to display prior analysis through overlay integration.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent introduces real-time imaging as an intermediary between prior static images and the current surgical field. The real-time images serve as a dynamic reference frame that accommodates tissue deformation and position changes, allowing prior analysis to be accurately mapped and displayed without being constrained by the rigidity of traditional co-registration methods.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Shape

If segmentation techniques are used to visualize organs, then three-dimensional visualization is achieved, but the ability to import prior analysis is limited

Engineering Contradiction:

Improvethree-dimensional visualizationVSAvoidprior analysis import capability

Core Design Contradiction:

ShapeVSLoss of information

Solution Approach 1:

The system merges segmentation-based three-dimensional visualization with real-time image fusion capabilities. By combining the spatial accuracy of segmented organ models with the dynamic positioning of real-time imaging, the system simultaneously achieves detailed three-dimensional visualization and the ability to import and display prior analysis on the same visual platform.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The visual化 system becomes multi-functional by integrating multiple capabilities: segmentation for three-dimensional organ visualization, real-time image acquisition for dynamic positioning, and prior analysis import for contextual information display. This universal platform eliminates the need to choose between different visualization approaches.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Loss of information

If static prior images are used for real-time visualization, then prior analysis can be displayed, but inaccuracies occur due to tissue manipulation and volume changes

Engineering Contradiction:

Improveprior analysis utilizationVSAvoidposition accuracy

Core Design Contradiction:

Loss of informationVSMeasurement precision

Solution Approach 1:

The system employs periodic real-time image acquisition during surgical procedures to continuously update the reference frame. This periodic capture of current tissue states allows the system to track and compensate for tissue manipulation and volume changes, maintaining position accuracy while displaying prior analysis throughout the procedure.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

Real-time imaging provides continuous feedback on actual tissue positions and deformations. This feedback is used to dynamically adjust the mapping and registration of prior analysis to the current surgical field, ensuring that position accuracy is maintained even as tissues are manipulated and volumes change during the procedure.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11138461B2Methods and systems for performing segmentation and registration of images using neutrosophic similarity scores

Publication Date: 2021.10.05 H LEE MOFFITT CANCER CENTER & RESEARCH INSTITUTE INC

AI summary

An example method for segmenting an object contained in an image includes receiving an image including a plurality of pixels, transforming a plurality of characteristics of a pixel into respective neutrosophic set domains, calculating a neutrosophic similarity score for the pixel based on the respective neutrosophic set domains for the characteristics of the pixel, segmenting an object from background of the image using a region growing algorithm based on the neutrosophic similarity score for the pixel, and receiving a margin adjustment related to the object segmented from the background of the image.