Dynamic Renal Scintigraphy CTT Estimation via Neural Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for quantitatively measuring cortical transit time (CTT) in dynamic renal scintigraphy suffer from lack of reproducibility and precision due to difficulties in determining the exact moment the tracer reaches calyceal cavities and sensitivity to perturbating factors, especially with low-resolution images.
Innovation Solution
A method using computer means to classify images based on kidney states before and after CTT, involving image normalization, 2D-convolution, and neural network classification to estimate CTT accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If mathematical methods with deconvolution are used to measure CTT, then quantitative measurement is achieved, but precision deteriorates due to low resolution of images and high sensitivity to perturbating factors
Solution Approach 1:
The patent segments the renal scintigraphy images into distinct regions: renal parenchyma region and calyceal cavity region. By dividing the image analysis into these separate segments, the method can independently track tracer dynamics in each region, avoiding the precision loss of global deconvolution methods while maintaining quantitative measurement capability.
Solution Approach 2:
The patent performs preliminary image processing including normalization and noise filtering before CTT measurement. It also pre-identifies the transition point between parenchymal impregnation and calyceal appearance, establishing a clear temporal marker that eliminates the need for sensitive deconvolution operations on the original low-resolution images.
2Ease of operation
If visual observation method is used to determine when tracer first reaches calyceal cavities, then simplicity is maintained, but measurement precision deteriorates due to difficulty in determining exact moment
Solution Approach 1:
The patent introduces an intermediary automated image analysis system that acts as a mediator between the simple visual observation method and precise quantitative measurement. This intermediary process includes automated region segmentation, tracer distribution tracking, and transition point detection, which maintains operational simplicity while dramatically improving measurement precision.
Solution Approach 2:
The patent replaces the manual visual observation mechanism with an automated computer-based image analysis system. This substitution eliminates human subjectivity and difficulty in determining the exact transition moment, while maintaining the overall simplicity of the workflow through automated processing.
3Measurement precision
If automated image analysis is implemented to improve CTT measurement precision, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent divides the complex image analysis task into manageable segments: normalization step, noise filtering step, parenchymal region identification, calyceal cavity identification, and transition point detection. This segmentation reduces the complexity of each individual processing step while achieving high overall precision.
Solution Approach 2:
The patent employs parameter changes in the form of image normalization and filtering parameters to simplify the analysis. By adjusting these parameters, the system enhances contrast and reduces noise, making the subsequent CTT measurement more precise without requiring complex algorithmic changes.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A method implemented by computer means for estimating the cortical transit time on a dynamic renal scintigraphy of at least one kidney of a patient, comprising the following steps: (a) obtaining at least one time sequence comprising at least two images from said scintigraphy, (b) classifying images of said at least one time sequence depending on a state of said kidney on the image, said state being before or after a cortical transit time of the kidney under consideration, (c) estimating the cortical transit time of each kidney from said classified images.