CC–MLO Breast ROI Correlation with Ensemble Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical imaging technologies struggle to accurately correlate regions of interest (ROIs) between cranial-caudal (CC) and medial-lateral-oblique (MLO) images of breast tissue, leading to imperfect and error-prone manual processes in identifying potential cancerous lesions.
Innovation Solution
A computing system utilizing machine learning models, including similarity and geo-matching models, to determine a joint probability of correlation between ROIs in CC and MLO images, providing a confidence level indicator to aid radiologists in identifying matching lesions across different image types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual correlation methods are used to identify lesions between CC and MLO images, then radiologists can review images, but the process is error-prone and time-consuming
Solution Approach 1:
The patent replaces manual mechanical review processes with automated machine learning models that compute joint probability of correlation. The ML model processes image data and outputs correlation results automatically, eliminating human error and time consumption while maintaining diagnostic accuracy.
Solution Approach 2:
The system enables self-service by automatically performing the correlation analysis without requiring radiologist intervention for each correlation determination. The ML model independently analyzes CC and MLO images, matches ROIs, and provides correlation probabilities autonomously.
2Reliability
If multiple ML models are combined to improve correlation accuracy, then diagnostic power increases, but system complexity increases
Solution Approach 1:
The patent merges multiple specialized ML models (similarity matching model, geo-matching model, and ensemble matching model) into a unified correlation system. Each model handles specific aspects of correlation analysis, and their combined output provides comprehensive correlation accuracy through complementary strengths.
Solution Approach 2:
The system segments the correlation analysis into distinct functional components: similarity matching for visual pattern recognition, geo-matching for spatial relationship analysis, and ensemble matching for integrated probability calculation. This segmentation allows each model to be optimized for its specific function while working together.
Data Source
AI summary
A method of correlating regions in an image pair including a cranial-caudal image and a medial-lateral-oblique image. Data from a similarity matching model is received by an ensemble model, the data including at least a matched pair of regions and a first confidence level indicator associated with the matched pair of regions. Data from a geo-matching model is received by the ensemble model, the data from the geo-matching model including at least the matched pair of regions and a second confidence level indicator. A joint probability of correlation is determined by the ensemble model based on evaluation of each of the first and second confidence level by the ensemble matching model, wherein the joint probability of correlation provides a probability that the region in each image correlates to the corresponding region in the other image. The joint probability of correlation is provided to an output device.