AI Lesion Correlation Across X-Ray and Ultrasound Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Clinicians face challenges in accurately correlating lesions identified in mammography or tomosynthesis with those in diagnostic ultrasound due to differences in patient position and imaging modality, leading to difficulties in locating smaller lesions.
Innovation Solution
A computing system utilizing artificial intelligence and machine learning to analyze x-ray and ultrasound images, providing a confidence level indicator for lesion correlation, and potentially employing electromagnetic or optical tracking to navigate to the target area during ultrasound.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If x-ray imaging (mammography/tomosynthesis) is performed with patient upright and breast compression to detect lesions, then lesion detection capability is improved, but correlation with ultrasound imaging becomes difficult due to position and compression differences
Solution Approach 1:
The system performs x-ray imaging with the patient upright and breast compression first to detect and mark the lesion location. This preliminary action establishes the target lesion position that will later guide the ultrasound imaging process, allowing the system to pre-calculate the transformation needed to locate the same lesion in the uncompressed, supine position during ultrasound.
Solution Approach 2:
The system introduces an intermediary coordinate transformation model that acts as a mediator between the x-ray imaging coordinate system (upright, compressed) and the ultrasound imaging coordinate system (supine, uncompressed). This transformation model, based on breast deformation physics, translates lesion coordinates from one modality to the other, resolving the correlation difficulty without requiring manual repositioning or re-imaging.
2Ease of operation
If diagnostic ultrasound is performed with patient supine and no breast compression to image tissue, then tissue visualization is improved, but locating previously identified x-ray lesions becomes challenging
Solution Approach 1:
The coordinate transformation model serves as an intermediary that translates lesion coordinates from the x-ray imaging system (where the lesion was precisely identified) to the ultrasound imaging system. This allows the ultrasound system to maintain its advantageous supine, uncompressed positioning for tissue visualization while still achieving precise lesion location through calculated coordinate transformation rather than manual search.
Solution Approach 2:
The system replaces the manual mechanical process of physically repositioning the patient and breast between imaging modalities with an automated computational coordinate transformation system. Instead of mechanically resetting the breast to a different position for each modality, the system uses software-based transformation to adapt coordinates, eliminating the need for physical repositioning and maintaining optimal positioning for each modality.
3Measurement precision
If imaging resolution is increased to detect smaller lesions, then early cancer detection is improved, but difficulty in locating these small lesions during ultrasound increases
Solution Approach 1:
The coordinate transformation model acts as an intermediary guide that provides precise computational coordinates for small lesions detected in x-ray imaging. This allows the ultrasound system to focus its search on specific transformed coordinate locations rather than manually scanning large areas, making it significantly easier to locate even very small lesions that are difficult to find through manual ultrasound examination alone.
Solution Approach 2:
The system replaces the manual visual search and physical navigation method with an automated computational coordinate transformation and guidance system. This substitution enables precise localization of small lesions through calculated coordinates rather than relying on the operator's manual search capability, dramatically improving the ease of detecting and measuring small lesions across different imaging 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
Enhances the ability to accurately locate lesions identified in x-ray imaging within ultrasound images, improving navigation and reducing the imaging area to be analyzed by providing a confidence level indicator.
Implementation Method 1
Ultrasound uses sound waves, typically produced by piezoelectric transducers, to image tissue in a patient
Data Source
AI summary
Methods and systems for identifying a region of interest in breast tissue utilize artificial intelligence to confirm that a target lesion identified during imaging the breast tissue using a first imaging modality (e.g. x-ray imaging) has been identified using a second imaging modality (e.g. ultrasound imaging). A computing system operating a lesion matching engine utilizes a machine learning classifier algorithm trained on cases of x-ray images and corresponding ultrasound images in which lesions were identified for further analysis. The lesion matching engine analyzes a target lesion identified with x-ray imaging and a potential lesion identified with ultrasound imaging to determine a likelihood that the target lesion is the same as the potential lesion. A confidence level indicator for the lesion match is presented on a display of a computing device to aid a healthcare provider in locating a lesion in breast tissue.