3D Time-Intensity Images for Lesion Severity in DCE MRI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical image processing methods for analyzing breast tumors using MRI generate only one-dimensional time-intensity curves (TICs), which provide a one-sided reference and cannot express the lesion region completely and accurately.
Innovation Solution
The method generates 3D first-stage and second-stage time-intensity images based on dynamic contrast enhanced (DCE) magnetic resonance images, using average pixel grayscale values and determining a time to peak to represent the change rate of blood supply intensity and reflect the severity level of the lesion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only one-dimensional time-intensity curves (TICs) are generated from DCE MRI data, then the processing method is simple, but the lesion region cannot be expressed completely and accurately
Solution Approach 1:
The patent transforms the one-dimensional TIC curve into two-dimensional time-intensity images by adding spatial dimensions. The method generates multiple time-intensity images from different time points of DCE MRI, where each image represents the lesion region at a specific time. This dimensional expansion allows comprehensive visualization of lesion characteristics including blood supply intensity, enhancement patterns, and temporal evolution, thereby achieving complete and accurate lesion expression.
2Loss of information
If one-dimensional TIC curves are used for breast tumor diagnosis, then the diagnostic process is straightforward, but the reference provided is one-sided and insufficient
Solution Approach 1:
The patent segments the DCE MRI data into multiple time points and generates separate time-intensity images for each time point. By dividing the continuous dynamic enhancement process into discrete temporal segments, the method preserves detailed information about lesion enhancement at different stages (early, intermediate, delayed phases). This segmentation approach prevents information loss while maintaining manageable processing complexity through systematic analysis of each time segment.
Solution Approach 2:
The patent adds temporal and spatial dimensions to transform one-dimensional TIC curves into two-dimensional time-intensity images. Each pixel in the time-intensity image represents a spatial location in the lesion, and the intensity value represents the signal enhancement at a specific time point. This dimensional transformation comprehensively captures lesion characteristics including heterogeneity, enhancement kinetics, and spatial distribution, providing complete lesion information for diagnosis.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Disclosed are an image processing method, apparatus and system, and an electronic device and a storage medium, relating to the technical field of medical image processing. The method comprises: acquiring DCE magnetic resonance images corresponding to multiple time points for the same detection target (200); respectively determining an average pixel grayscale value of images in the same lesion area in the DCE magnetic resonance image at each time point (210); determining a time to peak according to the average pixel grayscale value corresponding to each time point (220); and respectively generating, according to the DCE magnetic resonance image at each time point and the time to peak, a first-phase time-signal intensity image before the time to peak and a second-phase time-signal intensity image after the time to peak, wherein the first-phase time-signal intensity image and the second-phase time-signal intensity image are 3D images, and a pixel grayscale value of each point in the first-phase time-signal intensity image and the second-phase time-signal intensity image represents the rate of change in the degree of blood supply, and is used for reflecting the severity of a lesion (230).