Biomedical Image Registration via Automatic Landmark Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image registration methods for biomedical images, particularly in MRI and Mammography, face challenges in accurately tracking landmarks due to noise and movement artefacts, especially in soft tissues like the breast, leading to inaccurate registration and high false positive diagnoses.
Innovation Solution
A method using classification algorithms to determine valid landmarks by organizing characteristic parameters into vectors and training a classification algorithm with a database of known landmarks, which helps in identifying reliable features and reducing noise-related artefacts across different tissue types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If visual landmark identification by human operator is used, then landmark selection is flexible, but accuracy deteriorates due to operator skill dependence and difficulty in identifying landmarks in images without recognizable structures
Solution Approach 1:
The system performs automatic landmark identification and tracking without human operator intervention. The algorithm independently detects features, determines their characteristics, tracks their movement between images, and computes registration transformations, making the system self-sufficient and eliminating operator skill dependence while maintaining accuracy
Solution Approach 2:
The manual visual identification process is replaced with an automated computer-based algorithm that uses image processing techniques to detect and track landmarks. The system substitutes human cognitive and manual operations with computational methods, achieving consistent and reproducible landmark identification
2Extent of automation
If automatic landmark identification algorithms are used, then objectivity is improved, but reliability deteriorates due to noise and movement artefacts in biomedical images
Solution Approach 1:
The system employs iterative refinement where landmark detection results are evaluated, and the process is repeated with adjustments. The algorithm continuously refines landmark positions by comparing multiple images and using feedback from detection confidence measures, thereby improving reliability despite noise and artefacts in the images
Solution Approach 2:
The system performs preliminary processing steps including noise reduction, contrast enhancement, and feature preprocessing before landmark identification. By preparing the images in advance with optimized characteristics, the algorithm improves its ability to reliably detect and track landmarks even in challenging biomedical imaging conditions
3Stability of the object's composition
If compression panels are used to fix breast tissue, then fixation of contact areas is improved, but inner tissue movement is not controlled due to continued physiological motion from heartbeat and breathing
Solution Approach 1:
The system transitions from assuming static tissue positions to dynamically tracking landmark movements throughout the image acquisition sequence. By continuously monitoring and compensating for physiological movements in each frame, the algorithm maintains accurate registration despite ongoing internal tissue motion that compression panels cannot prevent
4Difficulty of detecting and measuring
If landmarks are concentrated at outer boundaries with high contrast, then detection is easier, but registration accuracy deteriorates due to insufficient coverage of inner tissue areas
Solution Approach 1:
The system applies different detection strategies and criteria to different regions of the image. Outer boundary areas with high contrast use one detection approach, while inner tissue areas use optimized parameters and algorithms suited for lower contrast regions. This localized adaptation ensures reliable landmark detection throughout the entire tissue volume, not just at boundaries
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In a method for registering biomedical images such as at least a first and a second digital or digitalized image or set of cross-sectional images of the same object, within the first image or set of images a certain number of landmarks, so called features are individuated by selecting a certain number of pixels or voxels. The position of each pixel or voxel selected as a feature is tracked from the first to the second image or set of images by determining the optical flow vector from the first to the second image or set of images for each pixel or voxel selected as a feature. Registration of the first and second images or set of images is carried out by applying the inverse optical flow to the pixels or voxels of the second image or set of images. The invention provides for an automatic trackable landmark selection step consisting in defining a pixel or voxel neighbourhood around each pixel or voxel of the first image or first set of cross-sectional images; for each target pixel or voxel determining one or more characteristic parameters which are calculated as a function of the parameters describing the appearance of the said target pixel or voxel and of each or a part of the pixels or voxels of the window and as a function of one or more characteristic parameters of either the numerical matrix or of a transformation of the said numerical matrix representing the pixels or voxels of the said window. The pixels or voxels coinciding with validly trackable landmarks are determined as a function of the said characteristic parameters of the target pixels or voxels.