Contrast-Enhanced Image Correction via Density Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In medical imaging, contrast-enhanced images acquired at different time frames show varying contrast agent densities due to degradation and dilution, making it difficult for clinicians to compare and analyze, especially in applications like cardiac imaging where time variability complicates the detection of physiological structures.
Innovation Solution
A computer-implemented method that selects a reference image frame with high contrast agent density, performs segmentation to identify a region of interest, and corrects contrast agent density differences in subsequent frames based on image intensity changes within this region, using non-linear transfer functions or histogram matching to align gray-value statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If contrast-enhanced images are acquired at different time frames, then the imaging sequence captures dynamic physiological processes, but the contrast agent density varies due to degradation and dilution making image comparison difficult
Solution Approach 1:
The patent creates a corrected copy of the original image sequence where contrast agent density is normalized across all frames. By generating corrected image frames that replicate the appearance of a reference frame with stable contrast density, the system enables reliable comparison without altering the original temporal dynamics of the imaging sequence.
Solution Approach 2:
The patent applies parameter transformation by changing the contrast agent density parameter across frames to match a reference frame. Through histogram matching and intensity normalization, the system transforms the density parameter while preserving other image characteristics, allowing consistent comparison across different time points.
2Extent of automation
If segmentation algorithms are applied to sequences with time-varying contrast agent density, then automated clinical information extraction is attempted, but errors occur due to detection of boundaries between regions of different agent density that do not correspond to physiological structures
Solution Approach 1:
The patent performs preliminary correction of contrast agent density variations before segmentation is applied. By pre-processing the image sequence to normalize contrast density across frames, the system eliminates spurious boundaries that would otherwise interfere with segmentation algorithms, ensuring that only true physiological boundaries are detected.
Solution Approach 2:
The patent creates corrected copies of images with uniform contrast agent density that serve as input for segmentation algorithms. These corrected images replicate the anatomical structures without the time-varying density artifacts, allowing segmentation to accurately identify physiological boundaries.
3Stability of the object's composition
If contrast agent density is allowed to vary naturally over time, then the natural physiological processes are captured, but the time variability complicates the detection of physiological structures
Solution Approach 1:
The patent selectively changes the contrast agent density parameter to match a reference frame while preserving other image parameters. Through intensity normalization and histogram matching, the system transforms only the density parameter to achieve stability, making physiological structures easier to detect without altering the underlying anatomy.
Solution Approach 2:
The patent introduces a reference frame as an intermediary standard against which all other frames are compared and corrected. This reference frame serves as a mediator that defines the target contrast density, allowing consistent detection of physiological structures across the entire image sequence.
Data Source
AI summary
A method and system for correcting a difference in contrast agent density in a sequence of contrast-enhanced image frames. A reference image frame is defined in the sequence of contrast-enhanced image frames, and segmentation is performed on the reference image frame to determine a location of a region of interest within the reference image frame. The region of interest is a region of the reference image frame that contains contrast agent. Other image frames in the sequence of contrast-enhanced images are corrected based on a difference in contrast agent density/image intensity in the region of interest relative to the region of interest in the reference image frame.