Radiation Energy Estimation for Stable Image Subtraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fluoroscopic images using energy subtraction become unstable due to fluctuations in radiation energy, particularly when the auto brightness control function adjusts the tube voltage, current, and pulse width, leading to overlapping references and unstable calibration curves.
Innovation Solution
An image processing apparatus and method that generates material characteristic images from multiple radiation images of different energies and estimates radiation energies by comparing these images over time, using a similarity index to optimize energy spectrum estimation and maintain stable energy subtraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If auto brightness control function is used to maintain constant radiation amount, then image quality stability is improved, but radiation energy fluctuation increases causing energy subtraction image instability
Solution Approach 1:
The system continuously monitors material characteristic images and uses feedback to estimate and correct radiation energy fluctuations. By comparing material characteristic images across different timings and using similarity indices, the system dynamically adjusts for energy variations caused by ABC function, maintaining stable energy subtraction results despite radiation energy fluctuations.
Solution Approach 2:
The patent changes the approach from directly controlling radiation parameters to monitoring and correcting material characteristic parameters. By estimating radiation energy based on material characteristic images and using similarity indices to detect changes, the system adapts to parameter variations without direct control, resolving the contradiction between ABC function stability and energy subtraction reliability.
2Measurement precision
If reference is placed in irradiation field to correct calibration curve, then measurement precision is improved, but reference may overlap with subject causing calibration failure
Solution Approach 1:
The patent extracts the reference function from physical reference objects placed in the irradiation field and transfers it to virtual references derived from material characteristic images. By using similarity indices to compare and estimate radiation energy from image data, the system eliminates the need for physical references that could overlap with subjects, maintaining measurement precision without the harmful overlap issue.
Solution Approach 2:
The system creates virtual copies of reference information through material characteristic images. Instead of using physical references, the patent generates reference data by analyzing and comparing material characteristic images across different timings, effectively copying the reference function into the image processing domain where it cannot conflict with the subject.
Data Source
AI summary
An image processing apparatus generates a material characteristic image based on a plurality of radiation images corresponding to a plurality of radiation energies, and estimates at least one of the plurality of radiation energies at a second timing using a plurality of material characteristic images including a material characteristic image generated based on a plurality of radiation images acquired at a first timing and a material characteristic image generated based on a plurality of radiation images acquired at the second timing later than the first timing.