Breast MRI Advisory Region Identification via Diffusion T2 Correlation
Solution Overview
Problem
Magnetic Resonance Imaging (MRI) techniques face challenges in accurately localizing abnormal tissue regions in breast tissue due to spatial distortions in echo planar diffusion-weighted images, making it difficult to determine the exact location of malignant tumors.
Innovation Solution
A medical system and method that combines echo planar diffusion-weighted MRI with fat-suppressed T2-weighted MRI to identify high diffusion rate regions and tissue regions, using image processing to correlate and segment these areas, providing advisory regions for further investigation, and calculating conductivity measures to classify potential cancerous areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If echo planar diffusion-weighted MRI is used to identify high diffusion rate regions, then abnormal tissue regions can be effectively identified, but spatial distortions occur making exact location uncertain
Solution Approach 1:
The patent introduces a second MRI sequence (T2-weighted or PD-weighted imaging) as an intermediary reference that provides accurate anatomical localization without the distortion artifacts. This intermediary image serves as a spatial reference frame to which the diffusion-weighted image data is mapped, thereby resolving the location uncertainty while preserving the abnormal tissue identification capability
Solution Approach 2:
The patent merges data from two different MRI sequences: the diffusion-weighted image (which identifies abnormal tissue through high diffusion rate regions) and the T2-weighted or PD-weighted image (which provides accurate spatial localization). By combining these complementary datasets through image coregistration and mapping, the system achieves both reliable abnormal tissue identification and precise spatial location
2Measurement precision
If two types of MRI images are combined and processed, then accurate localization of abnormal regions is achieved, but device complexity increases
Solution Approach 1:
The patent performs preliminary actions by acquiring both MRI sequences (diffusion-weighted and T2-weighted or PD-weighted) during the same imaging session with pre-established spatial correspondence. The images are acquired in a coordinated manner with matching anatomical coverage, which simplifies subsequent processing by ensuring proper alignment and reducing the complexity of image coregistration
Solution Approach 2:
The patent creates a simplified representation by mapping the diffusion-weighted image data onto the T2-weighted or PD-weighted image framework. Instead of processing both images equally, the system copies the abnormal tissue identification results from the diffusion-weighted image and overlays them on the anatomically accurate reference image, reducing processing complexity while maintaining precision
Data Source
AI summary
Disclosed herein is a method of medical imaging. The method comprises: receiving (200) an echo planar diffusion weighted magnetic resonance image (122) of a region of interest (309) descriptive of breast tissue; receiving (202) a fat suppressed T2 weighted magnetic resonance image (124) descriptive of the region of interest; segmenting (204) the echo planar diffusion weighted magnetic resonance image to identify high diffusion rate regions (128); segmenting (206) the fat suppressed T2 weighted magnetic resonance image to identify tissue regions (130); identifying (208) a portion of the tissue regions as advisory regions (134) by inputting the high diffusion rate regions and the tissue regions into an image processing module; and providing (210) the advisory regions as a segmentation of the fat suppressed T2 weighted magnetic resonance image.