Time-Axis Image Filtering for Angiography Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for displaying angiographic images using medical imaging apparatuses like X-ray CT and MRI fail to effectively create clear, three-dimensional images of contrasted blood vessels without errors from timing or location discrepancies and do not efficiently reduce the amount of contrast medium exposure, leading to suboptimal image quality and increased radiation exposure.
Innovation Solution
An image processing method that extracts and constructs two- or three-dimensional images by filtering along the time axis, identifying maximum and minimum pixel values of contrast medium presence, and applying weighting techniques to reduce noise and enhance image clarity without increasing contrast medium usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Digital Subtraction Angiography (DSA) is used to extract contrasted images by taking images before and after contrast medium injection, then the contrasted blood vessel images can be obtained, but uneven results occur due to errors in imaging timings or imaging locations
Solution Approach 1:
The patent applies preliminary action by extracting the maximum pixel values along the time axis from multiple images taken at different time points before performing the subtraction operation. This preprocessing step ensures that the contrasted blood vessel information is maximized and aligned properly, preventing uneven results due to timing or location errors in the original images. By preparing the data in advance with maximum value extraction, the subsequent subtraction produces more reliable and consistent results.
2Loss of information
If multiple images are taken to capture contrast medium passage through blood vessels, then the dynamic process can be visualized, but the amount of contrast medium required increases
Solution Approach 1:
The patent implements continuity of useful action by processing multiple images taken during the continuous passage of contrast medium through blood vessels. Instead of requiring separate contrast medium injections for different views or time points, the system continuously captures images as the contrast medium naturally flows through the vasculature, extracting maximum values along the time axis to preserve the dynamic process information while using the same contrast medium dose throughout the sequence.
3Measurement precision
If conventional angiography methods are used to obtain clear blood vessel images, then image quality can be maintained, but radiation exposure to the subject increases
Solution Approach 1:
The patent applies the extraction principle by isolating and extracting only the relevant contrasted blood vessel information from multiple images through maximum value extraction along the time axis and subtraction operations. This allows the system to obtain clear blood vessel images by extracting the essential information needed, rather than requiring additional radiation exposure from repeated imaging or higher dose single images. The extraction process concentrates the useful information while discarding redundant data that would require additional radiation to capture.
4Adaptability or versatility
If manual region extraction is performed by operators to analyze specific areas, then customization is possible, but operator variability affects accuracy and time consumption increases
Solution Approach 1:
The patent applies self-service by implementing automated processing that performs maximum value extraction along the time axis and subtraction operations without requiring manual operator intervention for each image or region. The system automatically identifies and extracts the contrasted blood vessel information, eliminating operator variability and reducing time consumption while maintaining the ability to analyze specific regions through programmatic control rather than manual selection.
Data Source
AI summary
An image processing method comprises: an inputting step of inputting image data which is obtained by imaging a subject for a predetermined period of time with a medical imaging apparatus and is arranged in time series; an extracting along the time axis step of extracting pixels which satisfy a predetermined condition along the time axis from all the pixels arranged in time series for each pixel coordinate position with respect to each pixel in the image data; and a constructing step of constructing a two-dimensional or three-dimensional image based on the pixels extracted along the time axis in the extracting along the time axis step.