Medical Image Processing Apparatus for DWI-FLAIR Mismatch Quantification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting the onset time of cerebral infarction using DWI/FLAIR mismatch phenomenon struggle to quantitatively distinguish between affected and unaffected sides in cases of bilateral infarctions, leading to qualitative evaluations by doctors.
Innovation Solution
A medical image processing apparatus that identifies a first area related to a lesion in a DWI image, determines its bilateral or unilateral nature, and calculates a mismatch index using feature amounts from corresponding areas in a FLAIR image, enabling accurate quantitative evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the DWI/FLAIR mismatch phenomenon is quantified using the affected side area and unaffected side area in the FLAIR image, then the accuracy of cerebral infarction phase determination is improved, but in cases of bilateral infarction the method cannot distinguish between affected and unaffected sides leading to qualitative evaluation
Solution Approach 1:
The patent applies asymmetry by determining whether the affected side area is located on the left or right side of the brain, and accordingly selecting the symmetric area on the opposite side. This asymmetric approach based on lesion location enables proper identification of unaffected areas even in bilateral infarction cases, resolving the limitation of conventional symmetric comparison methods.
Solution Approach 2:
The patent segments the brain into left and right hemispheres and further divides affected areas into multiple regions. By segmenting the affected side area into plural regions and determining their respective locations, the system can properly match each affected region with its corresponding unaffected region, enabling quantitative evaluation even when multiple infarcts are present on both sides.
2Productivity
If doctors qualitatively determine the DWI/FLAIR mismatch phenomenon in their own mind for bilateral infarction cases, then the examination can be completed, but the measurement precision and objectivity are reduced
Solution Approach 1:
The patent implements self-service by enabling the image processing system to automatically perform the mismatch evaluation that previously required doctor's subjective judgment. The system automatically determines affected side area location, identifies corresponding unaffected areas, calculates feature amounts, and computes mismatch indices without requiring manual qualitative assessment, thereby maintaining both productivity and objectivity.
Solution Approach 2:
The patent changes the evaluation from qualitative to quantitative by introducing specific parameters: affected side area location (left/right), symmetric area identification, feature amounts (mean pixel values), and mismatch indices. These parameter changes enable objective numerical comparison even in complex bilateral cases, replacing subjective qualitative determination.
3Measurement precision
If the affected side area is mapped onto the FLAIR image and symmetric area is decided based on line symmetry with the center line, then the DWI/FLAIR mismatch can be quantified in unilateral cases, but this method fails when infarcts occur on both left and right sides
Solution Approach 1:
The patent applies dynamics by making the symmetric area selection adaptive rather than fixed. The system dynamically determines the symmetric area based on the actual location of the affected side area - if the affected area is on the left, the symmetric area is on the right, and vice versa. This dynamic adaptation enables proper comparison in both unilateral and bilateral infarction cases.
Solution Approach 2:
The patent achieves universality by creating a method that works for both unilateral and bilateral infarction cases. By incorporating affected side location determination and adaptive symmetric area selection, the system provides a unified quantitative evaluation approach that handles various infarction patterns (unilateral, bilateral, multiple regions) through the same procedural framework.
Data Source
AI summary
A medical image processing apparatus includes processing circuitry that identifies a first area with a lesion in a site of a subject based on a first medical image generated by first imaging of the site; determines whether the first area is bilateral or unilateral relative to the site, based on a position of the first area; upon determining that as bilateral, decides, as a second area, a normal portion in the site in a second medical image generated by different, second imaging; upon determining that as unilateral, decides, in the second medical image, as the second area, an area line symmetric to the first area about a centerline of the site; calculates, in the second medical image, first and second feature amounts of the respective first and second areas; and calculates an index representing mismatch between the first and second medical images, based on the first and second feature amounts.