MRI Segmentation Using Reference Grayscale Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic Resonance Imaging (MRI) image segmentation is hindered by variations in image contrast due to changes in acquisition protocols, leading to inaccurate tissue type identification and segmentation across different MRI systems and protocols.
Innovation Solution
Incorporating reference grayscale value data into the image segmentation algorithm, which can be generated through modeling, phantom measurements, or statistical analysis, to adjust and improve the segmentation process, allowing for more accurate identification of anatomical structures and tissues regardless of the imaging protocol or system used.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic segmentation algorithms are used to identify anatomical structures, then productivity is improved, but measurement precision deteriorates due to variations in grayscale values across different acquisition protocols
Solution Approach 1:
The patent transforms grayscale values into protocol-independent appearance values by applying calibration functions that account for acquisition protocol parameters (TR, TE, flip angle). This parameter transformation allows the segmentation algorithm to maintain high measurement precision across different protocols while preserving automation benefits.
Solution Approach 2:
The patent introduces appearance values as an intermediary representation between raw grayscale values and segmentation results. These appearance values serve as a standardized intermediate form that eliminates protocol-dependent variations, enabling accurate automated segmentation without requiring protocol-specific algorithm adjustments.
2Measurement precision
If segmentation algorithms are adapted to different MRI systems and protocols, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent creates a universal appearance value representation that works across all MRI systems and protocols. The calibration function uses standard acquisition parameters (TR, TE, flip angle) to transform any protocol's grayscale values into a common appearance value space, eliminating the need for protocol-specific algorithm versions and reducing overall system complexity.
Solution Approach 2:
By changing the parameter space from protocol-dependent grayscale values to protocol-independent appearance values, the patent simplifies the segmentation algorithm. The algorithm operates on standardized appearance values rather than raw grayscale values, making it universally applicable without requiring complex protocol-specific adaptations.
3Measurement precision
If reference grayscale value data is incorporated into the segmentation algorithm, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent transforms the segmentation approach by changing from direct grayscale value comparison to appearance value comparison. The reference data is stored as appearance values (protocol-independent) rather than raw grayscale values, and the calibration function automatically handles the transformation. This parameter change improves precision while keeping the algorithm relatively simple.
Solution Approach 2:
The patent performs preliminary transformation of grayscale values to appearance values using calibration functions before the actual segmentation process. This preliminary action prepares the data in a standardized format, allowing the segmentation algorithm to operate with high precision without requiring complex real-time adjustments during execution.
Data Source
AI summary
The invention provides for a medical imaging system (100, 300, 400) comprising a memory (110) containing machine executable instructions (120) and a processor (106). Execution of the machine executable instructions cause the processor to: receive (200) a magnetic resonance image (300), receive (202) meta data descriptive of the magnetic resonance image, wherein the metadata comprises reference gray scale value data (124) for two or more tissue types; and segment (204) the magnetic resonance image using an image segmentation algorithm (126) that uses the reference gray scale value data.