Mammogram Normalisation With Local Ratio Maps for BAC Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CAD systems struggle to consistently differentiate and segment breast arterial calcifications (BACs) in mammograms due to variations in imaging modalities and equipment, leading to high false positive rates and reduced sensitivity.
Innovation Solution
A method and system that normalizes mammograms by transforming pixel values to be independent of imaging conditions, using contrast-stretching algorithms, local maximum transforms, and ratio maps to enhance visibility and segment BACs effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional CAD systems are used to detect BACs in mammograms, then automated analysis can be performed, but false positive rates increase and sensitivity decreases due to imaging condition variations
Solution Approach 1:
The patent transforms pixel values from absolute intensity values to relative magnitude values representing the relationship between each pixel and its neighboring pixels. This parameter transformation makes the features independent of imaging conditions such as x-ray tube voltage, radiation dose, and filter material, thereby improving detection reliability while maintaining automation
Solution Approach 2:
The patent introduces an intermediary processing step that creates a normalized representation of mammogram images by computing relative magnitudes. This intermediary form serves as a bridge between raw images and CAD analysis, eliminating the negative effects of imaging condition variations before detection algorithms are applied
2Productivity
If raw mammograms with varying imaging conditions are analyzed directly, then processing time is reduced, but detection consistency across different modalities and manufacturers deteriorates
Solution Approach 1:
The patent applies a lightweight parameter transformation that computes relative magnitudes between pixels and their neighbors. This transformation requires minimal computation compared to traditional preprocessing pipelines, maintaining high processing speed while achieving consistent detection results across different imaging modalities and manufacturers
3Measurement precision
If contrast enhancement is applied to improve BAC visibility, then detection sensitivity improves, but false positive rate increases due to enhanced noise and artifacts
Solution Approach 1:
The patent transforms the image representation from absolute pixel intensities to relative magnitudes between pixels and their neighboring pixels. This parameter change inherently enhances contrast for BAC detection while being insensitive to uniform noise and artifacts, thereby improving sensitivity without increasing false positives
Solution Approach 2:
The patent computes relative magnitudes locally for each pixel based on its relationship with neighboring pixels. This local approach enhances local features such as BACs while maintaining robustness against global noise and artifacts, achieving improved detection precision without generating additional false positives
Data Source
Figure 1(a)~1(d)
Figure 2(a)~2(d)
Figure 3(a)~3(d)
AI summary
The present invention relates to a method and system for the transformation of raw mammograms to normalised presentation and where the pixel values are independent of imaging conditions. The performed method includes: contrast enhancement, for improved visibility of the breast tissue composition, whereby a region of the breast is segmented and a contrast-stretching algorithm applied to the segmented region to preferably create an enhanced raw image or mammogram; local 'maximum' transform, whereby a 2-dimentional first filter is designed to extract the maximum pixel value from a region of interest (ROI) to preferably create a local maximum image or map; ratio map derivation, whereby the pixel value of the ratio map measures a relative response of the said pixel to its local maximum thus capturing the difference between breast composition regardless of mammogram variations.