Automated Cerebral Blood Volume Map Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for automated tumor grading in brain tumors using cerebral blood volume maps are prone to user-induced bias and require manual normalization, which is time-consuming and subjective, potentially leading to inaccurate tumor grading due to uncorrected blood volume maps.
Innovation Solution
The method employs k-class cluster analysis, specifically k-means and fuzzy c-means, to segment blood vessels from dynamic susceptibility contrast MRI images, generating vessel masks that reduce false positives and allow for automated tumor grading with similar diagnostic accuracy to manual methods, by normalizing perfusion-related maps using relative blood volume and other curve parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual normalization of cerebral blood volume maps is performed by experienced neuroradiologists, then diagnostic accuracy is maintained, but time consumption increases and user-induced bias is introduced
Solution Approach 1:
The system performs automated normalization of cerebral blood volume maps using algorithm-based methods that self-correct for contrast agent extravasation effects. The computer automatically identifies tumor regions and calculates normalized rCBV values without requiring manual intervention by neuroradiologists, thereby eliminating user-induced bias and significantly reducing time consumption while maintaining diagnostic accuracy
Solution Approach 2:
The manual mechanical process of normalization by neuroradiologists is replaced with an automated computational system. The invention uses computer-based algorithms to perform normalization calculations, substituting the manual mechanical/visual assessment process with automated image processing and mathematical computations that eliminate human variability and time constraints
2Productivity
If automated methods for segmenting tumors from MR images are used, then productivity increases, but measurement precision may deteriorate due to uncorrected blood volume maps
Solution Approach 1:
The automated segmentation system incorporates self-correction capabilities by automatically detecting and correcting for contrast agent extravasation effects within the algorithm. The system performs both tumor segmentation and rCBV normalization in an integrated automated workflow, ensuring that productivity gains do not compromise measurement precision through the inclusion of correction algorithms
Solution Approach 2:
The automated system performs multiple functions including tumor segmentation, blood volume map normalization, and extravasation correction within a single integrated workflow. This multi-functional approach ensures that productivity improvements through automation do not sacrifice measurement precision, as all corrections are applied automatically in the same system that performs segmentation
3Measurement precision
If contrast agent extravasation correction is applied to rCBV maps, then measurement precision improves, but device complexity increases
Solution Approach 1:
The system performs automatic detection and correction of contrast agent extravasation effects through built-in algorithms that analyze the temporal characteristics of contrast enhancement. By making the correction process automatic and integrated into the normalization workflow, the system improves measurement precision without requiring complex manual intervention or additional specialized equipment
Solution Approach 2:
The invention corrects rCBV measurements by analyzing changes in signal intensity over time and applying mathematical corrections based on detected extravasation parameters. By changing the processing parameters and applying correction factors algorithmically, the system improves measurement precision while keeping the overall device complexity manageable through software-based solutions rather than hardware complexity
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
This approach provides accurate and automated tumor grading with reduced user dependence, improving diagnostic accuracy and efficiency by removing false positives and introducing minimal false negatives, and allows for the differentiation of tumor vascularity at the capillary level, achieving comparable or better results than manual grading.
Implementation Method 1
The method employs k-class cluster analysis, specifically k-means and fuzzy c-means, to segment blood vessels from dynamic susceptibility contrast MRI images
Implementation Method 2
perfusion imaging or DCE imaging refers to techniques such as T 1 -, T 2 -, or T 2 *- weighted imaging, such as preferably dynamic susceptibility contrast (DSC) imaging
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
The invention relates to automated normalization of cerebral blood volume maps, from perfusion MR images such as dynamic susceptibility contrast (DSC) images. According to the invention, noisy pixels are identified and excluded, tumor pixels are identified and excluded, and pixels thought to represent normal-appearing brain tumor are identified and used to assess a mean reference CBV value for each DSC-slice. This normalization procedure is particularly useful for normalizing CBV maps in DCS-based glioma images.