Multi-modality Breast Imaging Deformation Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-modality breast imaging techniques face challenges in accurately merging information from breast MRI and Digital Breast Tomosynthesis (DBT) due to differences in patient positioning and compression levels, making it difficult to visually combine features from both modalities effectively.
Innovation Solution
A system comprising shape model constructing subsystems for natural and compressed breast images, a deformation estimating subsystem using an elastic deformation model and landmark identification, which estimates a volumetric deformation field to map between images, allowing for accurate registration and fusion of breast images regardless of compression levels, and providing additional information about the skin surface and tissue volume.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If breast compression is applied for DBT imaging, then image quality and tissue compression are improved, but the natural shape of the breast is altered making registration with MRI difficult
Solution Approach 1:
The system performs preliminary actions by acquiring MRI images of the breast in its natural, uncompressed state before DBT imaging with compression. Shape models are constructed from these preliminary MRI images to capture the breast's natural geometry. This allows the registration process to account for shape transformations caused by compression, enabling accurate mapping between compressed DBT images and uncompressed MRI images.
2Measurement precision
If compression paddle positions are used for deformation simulation, then registration accuracy is improved, but the system complexity and requirement for exact position data increase
Solution Approach 1:
Instead of directly using complex compression paddle position data and contact mechanics models, the system creates simplified shape models that copy the essential geometric information from MRI images. These shape models serve as surrogate representations that capture the breast's deformation characteristics without requiring exact replication of the complex compression mechanics, thereby reducing system complexity while maintaining registration accuracy.
3Measurement precision
If landmark-based registration is used, then corresponding features can be identified, but reliable landmarks may not be visible in both modalities
Solution Approach 1:
The system introduces shape models as intermediary representations that bridge the gap between MRI and DBT modalities. These shape models are constructed from MRI images and transformed according to compression characteristics, serving as a mediator that enables correspondence identification. By comparing the transformed shape model with the compressed DBT images, the system can identify corresponding anatomical features even when traditional landmarks are not visible in both modalities.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the reliability and accuracy of multi-modality breast imaging by enabling the registration of images with varying compression levels without requiring exact compression paddle positions, improving the identification of anatomical features and facilitating the fusion of features from different imaging modalities.
Implementation Method 1
a deformation estimating sub-system for estimating a volumetric deformation field defining a mapping between the first image and the second image on the basis of the shape models and an elastic deformation model of the breast
Implementation Method 2
a second shape model constructing sub-system for constructing a second shape model of the breast as represented in a second image, in which the breast is compressed by using a compression paddle
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system for multi-modality breast imaging comprises a first shape model constructing sub-system (1) for constructing a first shape model of the breast as represented in a first image (9), in which the breast has its natural shape, a second shape model constructing sub-system (2) for constructing a second shape model of the breast as represented in a second image (10), in which the breast is compressed by using a compression paddle, and a deformation estimating sub-system (3) for estimating a volumetric deformation field (12) defining a mapping between the first image (9) and the second image (10) on the basis of the shape models and an elastic deformation model (11) of the breast, the deformation estimating sub-system (3) being arranged to estimate the volumetric deformation field (12) on the basis of a first tissue surface of the breast in the first image (9) and a second tissue surface of the breast in the second image (10).