Breast Arterial Calcification Detection via Adaptive Polynomial Pruning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting calcifications in mammograms struggle to accurately construct global calcified arterial structures and quantify their presence, which is crucial for predicting heart disease risk, as they often lead to false identifications and fail to effectively remove global arterial structures.
Innovation Solution
A novel method involving image processing techniques that include de-noising mammograms, computing points of interest, constructing mesoscale regions, computing connectivity, selecting branches based on physical properties, pruning branches using an adaptive polynomial fit, and growing chains to represent macroscopic calcified arterial structures, enabling accurate detection and quantification of calcified arterial structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated methods are used to detect calcifications in mammograms, then detection speed and objectivity are improved, but false identifications and inability to accurately construct global arterial structures occur
Solution Approach 1:
The method segments the detection process into distinct phases: noise removal from mammograms, detection of individual calcification points, construction of mesoscale regions, and assembly of global arterial structures. This segmentation allows each phase to be optimized independently, improving both speed and accuracy.
Solution Approach 2:
Mesoscale regions serve as an intermediary structure between individual calcification points and global arterial structures. These mesoscale regions act as building blocks that bridge the gap between point detections and complete arterial structure construction, resolving the contradiction by providing a structured intermediate representation.
2Loss of time
If simple detection methods are used, then processing time is reduced, but false positives increase and arterial structures cannot be accurately pruned
Solution Approach 1:
The method performs preliminary noise removal and mesoscale region construction before final arterial structure assembly. By preparing data in advance and organizing it into mesoscale regions, the subsequent pruning and identification steps can proceed efficiently with high precision, reducing overall processing time while maintaining accuracy.
Solution Approach 2:
The method applies different processing qualities to different scales: noise removal at the pixel level, mesoscale region construction at the local calcification level, and global arterial structure assembly at the macro level. This multi-scale approach with differentiated quality requirements optimizes both speed and precision at each level.
3Measurement precision
If detailed processing is applied to remove noise and prune branches, then detection accuracy is improved, but computational complexity increases
Solution Approach 1:
The computational process is segmented into modular steps: noise removal, point detection, mesoscale region construction, connectivity computation, and branch pruning. Each module handles a specific aspect of the problem, making the overall complex task manageable and allowing for optimized implementation of each segment.
Solution Approach 2:
The method applies partial processing at different scales rather than full processing everywhere. Mesoscale regions are constructed only where calcifications are present, and branch pruning is applied selectively to identified arterial structures, reducing unnecessary computational complexity while maintaining detection precision.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method includes receiving an image from a mammogram, removing noise from the image thereby creating a de-noised image, computing a point of interest on the de-noised image, creating a mesoscale region of interest on the de-noised image, computing a connectivity for the mesoscale region of interest, identifying a connected component using the computed connectivity, where the connected component represents a branch of a global curvilinear structure, selecting a set of branches based on a physical property for each branch of the global curvilinear structure, pruning each branch based on an error-tolerant, adaptive polynomial fit, identifying remaining regions of interest in each pruned branch, and growing a chain formed by remaining points of interest included in the remaining regions of interest, where the chain represents a macroscopic, global curvilinear calcified arterial structure. In an aspect of the invention, the quantitation of the calcified arterial structures may be used as a biomarker for risk stratification of heart disease.