Multi-Energy Radiation Image Processing for Component Separation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing energy subtraction processing methods using two radiation images with different energy distributions fail to accurately separate multiple components in a subject due to variations in the composition of soft tissues, such as fat and muscle, leading to incomplete separation of components like bone and soft parts.

Innovation Solution

A radiation image processing device and method that utilizes n radiation images with different energy distributions to derive characteristics and thicknesses of components, enabling accurate separation of n components by deriving attenuation coefficients and body thicknesses, enhancing component images individually.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If energy subtraction processing is performed using two radiation images with different energy distributions, then bone parts and soft parts can be separated, but multiple components (n components) cannot be accurately separated because only n component images can be obtained from n radiation images

Engineering Contradiction:
Improvecomponent separation accuracyVSAvoidnumber of radiation images required
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the parameter of radiation energy distribution by using n-1 types of radiation with different energy distributions. By deriving attenuation coefficients at multiple energy levels and using body thickness information, the system can separate n components (including soft parts, bone parts, and artificial objects) from only n-1 radiation images, achieving accurate multi-component separation without requiring n images for each component

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces body thickness as an additional dimension of information to resolve the component separation problem. By combining attenuation coefficient data from multiple energy levels with body thickness measurements, the system creates a multi-dimensional parameter space that enables separation of n components using only n-1 radiation images, effectively adding a dimensional constraint to the inverse problem

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If the attenuation coefficient of the soft part is not obtained in accordance with the ratio between fat and muscle, then accurate separation of multiple components cannot be achieved

Engineering Contradiction:
Improvesoft part component separation accuracyVSAvoidattenuation coefficient measurement difficulty
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent segments the soft part into multiple components (fat and muscle) and derives their respective attenuation coefficients separately. By dividing the soft part region and calculating attenuation coefficients for each component based on their different energy-dependent attenuation characteristics, the system can accurately determine the composition ratio and achieve precise separation of soft tissue components

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses feedback by iteratively deriving attenuation coefficients at multiple energy levels and comparing the derived body thickness with measured body thickness. This feedback loop allows the system to adjust and refine the attenuation coefficients of soft tissue components until the calculated and measured values converge, ensuring accurate determination of fat and muscle ratios

Inventive Principle:
Principle #23Feedback

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Accurately separates multiple components in a subject by using n radiation images, allowing for precise differentiation of components like soft parts and bone parts, as well as artificial objects, through enhanced image processing.

Implementation Method 1

the attenuation amount of the transmitted radiation differs in accordance with the substance constituting the subject

Methodology Applied
Scientific EffectAttenuation of radiation: Absorption (EM radiation)

Data Source

PatentEP4616810A1Radiation image processing device, radiation image processing method, and radiation image processing program
Publication Date: 2025.09.17 FUJIFILM CORP
  • EP4616810A1 patent drawingFigure 1~2
  • EP4616810A1 patent drawingFigure 3~4
  • EP4616810A1 patent drawingFigure 5

AI summary

A processor is configured to: acquire first to (n - 1)th (n ≥ 3) radiation images acquired by imaging a subject, which includes a first component consisting of a plurality of compositions and second to nth components each consisting of a single composition, with n - 1 types of radiation having different energy distributions; derive a characteristic of the first component in at least a region of the subject in the first to (n - 1)th radiation images; acquire a body thickness of the subject; derive thicknesses of the first to nth components by using the body thickness, the characteristic of the first component, and the first to (n - 1)th radiation images; and derive first to nth component images in which the first to nth components are enhanced, respectively, based on the thicknesses of the first to nth components.