3D Image Cross-Section Correlation Without Deformation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for displaying corresponding cross-sectional images from differently deformed three-dimensional medical images fail to maintain shape differences, requiring significant processing time and resources to deform images into identical shapes, and often cannot accurately generate corresponding cross-sectional images.

Innovation Solution

An image processing apparatus that calculates an approximate plane for a corresponding cross-section in one three-dimensional image based on deformation conditions of another image, allowing for the generation and display of corresponding cross-sectional images while maintaining shape differences between images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional methods deform images into identical shapes to display corresponding cross-sectional images, then the images can be compared, but significant processing time and resources are required

Engineering Contradiction:

Improveaccuracy of corresponding cross-sectional image generationVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the necessary deformation information (deformation amount and direction) from the full image deformation process. Instead of deforming entire images into identical shapes, the system extracts correlation information that correlates cross-sections between non-deformed and deformed three-dimensional images, significantly reducing processing requirements while maintaining accuracy in generating corresponding cross-sectional images

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary calculation of deformation amounts and directions between non-deformed and deformed three-dimensional images before generating cross-sectional images. By pre-calculating the correlation information that maps corresponding points between deformed and non-deformed states, the system avoids time-consuming post-processing deformation operations

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If conventional methods deform images into identical shapes to display corresponding cross-sectional images, then the images can be compared, but device complexity and processing resources increase

Engineering Contradiction:

Improveaccuracy of corresponding cross-sectional image generationVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces complex mechanical image deformation operations with computational correlation calculations. Instead of physically transforming image data through deformation algorithms, the system uses mathematical correlation information to map corresponding points between deformed and non-deformed images, substituting mechanical processing with efficient computational methods

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If physician manually estimates lesion position based on posture change deformation, then ultrasonic imaging can be performed, but the estimated position greatly differs from actual lesion position

Engineering Contradiction:

Improveease of lesion position estimationVSAvoidaccuracy of lesion position

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent establishes a feedback loop between non-deformed and deformed three-dimensional images by calculating correlation information that maps corresponding points. This feedback mechanism allows the system to accurately determine how lesions and other anatomical features have shifted due to posture changes, enabling precise localization without relying on manual physician estimation

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent replaces manual physician estimation with automated computational analysis. By substituting human visual assessment and manual positioning with algorithmic correlation calculations between deformed and non-deformed images, the system objectively determines lesion positions with higher precision, eliminating the large errors associated with manual estimation

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8917924B2Image processing apparatus, image processing method, and program

Publication Date: 2014.12.23 CANON KK

AI summary

An image processing apparatus includes a correlating unit configured to acquire correlation information that correlates a first three-dimensional image of a target object with a second three-dimensional image of the target object, and a corresponding cross-sectional image generation unit configured to generate a corresponding cross-sectional image of one of the first three-dimensional image and the second three-dimensional image, if a cross section is set on the other of the first three-dimensional image and the second three-dimensional image, based on the correlation information.