Brain Image Registration via Landmark Detection and Multi-Stage Transformation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for registering brain images with standard images are cumbersome for users, prone to inaccurate anatomical structure identification, and may unnaturally deform anatomical structures due to high degrees of freedom in affine transformations.

Innovation Solution

A medical image processing apparatus that detects specific reference landmarks (left eye, right eye, diencephalon, fornix, corpus callosum, left hippocampus, and right hippocampus) using template matching, followed by similarity transformation and nonlinear transformation for accurate registration between brain images and standard images, ensuring anatomical structures are not deformed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If affine transformation is used for registration between brain images and standard images, then the registration can be performed with high flexibility, but anatomical structures may be unnaturally deformed

Engineering Contradiction:
Improveregistration flexibilityVSAvoidanatomical structure integrity
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent divides the registration process into multiple stages: rigid transformation for global alignment, affine transformation for regional adjustment, and non-linear transformation for fine-tuning. Each stage operates with specific constraints to preserve anatomical integrity while achieving necessary flexibility.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different transformation types to different regions of the brain image. Rigid transformation is applied globally, affine transformation to specific regions needing adjustment, and non-linear transformation locally for fine-tuning, with each region's transformation constrained to preserve anatomical plausibility.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If manual marker setting is required for registration, then registration accuracy can be improved, but user work burden increases significantly

Engineering Contradiction:
Improveregistration accuracyVSAvoiduser work burden
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent implements automatic landmark detection using template matching and automated feature extraction algorithms. The system automatically identifies anatomical landmarks such as the corpus callosum, ventricles, and cortical surfaces without requiring manual user input, thereby reducing work burden while maintaining accuracy through algorithmic optimization.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical marker placement with automated computational methods including template matching, feature detection, and algorithmic registration. This substitution eliminates the need for physical marker setting while achieving comparable or superior registration accuracy through digital image processing techniques.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If high degree of freedom transformation is applied for registration, then adaptation to individual brain variations is improved, but anatomical structures become deformed

Engineering Contradiction:
Improveindividual variation adaptationVSAvoidanatomical structure shape
Core Design Contradiction:
Adaptability or versatilityVSShape

Solution Approach 1:

The patent employs a dynamic, multi-stage transformation approach where the type and extent of transformation applied at each stage depends on the specific anatomical region and the degree of variation detected. The system adapts the transformation parameters dynamically based on anatomical constraints, allowing high flexibility where appropriate while preventing deformation where anatomical integrity must be preserved.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS11315263B2Medical image processing apparatus, medical image processing method, and medical image processing program
Publication Date: 2022.04.26 FUJIFILM CORP
  • US11315263B2 patent drawing
  • US11315263B2 patent drawing
  • US11315263B2 patent drawing

AI summary

A medical image processing apparatus having a processor configured to detect at least four reference landmarks among the left eye, the right eye, the diencephalon, the fornix, the corpus callosum, the left hippocampus, and the right hippocampus from a brain image, performs first registration including registration by similarity transformation using reference landmarks between the brain image and a standard brain image, and perform second registration by nonlinear transformation between the brain image and the standard brain image after the first registration.