CT Projection Image Synchronization via Spectral Decomposition

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image reconstruction techniques for CT devices assume a certain periodicity in subject movement, leading to inaccuracies in reconstructing images of organs like the lungs and heart.

Innovation Solution

An information processing system that acquires waveform data from CT projection images, applies spectral decomposition to determine appropriate projection images for reconstruction without assuming a specific periodicity, using singular spectrum analysis to differentiate between exhalation/inhalation and diastole/systole.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If periodicity assumption is applied to subject movement in image reconstruction, then processing can be simplified, but reconstruction accuracy deteriorates when periodicity changes during measurement

Engineering Contradiction:

Improveprocessing complexityVSAvoidreconstruction accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies dynamics by making the periodicity parameter adjustable rather than fixed. The system detects actual periodicity from the projection images and adapts the reconstruction process to match the detected periodicity, allowing the system to respond to changes in subject movement patterns during measurement.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the periodicity parameter based on detected subject movement characteristics. By detecting periodicity from the projection images and using this detected periodicity in reconstruction, the system dynamically adjusts parameters to match actual measurement conditions, resolving the contradiction between simplified processing and accurate reconstruction.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If waveform decomposition is applied to detect periodicity, then reconstruction accuracy improves, but processing complexity increases

Engineering Contradiction:

Improveperiodicity detection accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary waveform decomposition and periodicity detection before the actual image reconstruction process. By detecting periodicity in advance from the projection images and preparing the decomposition results beforehand, the system enables accurate reconstruction without adding excessive complexity during the main processing phase.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250308097A1Information processing system, information processing apparatus, information processing method, and program

Publication Date: 2025.10.02 RIGAKU CORP

AI summary

According to an aspect of the present disclosure, an information processing system for processing a plurality of projection images taken by a CT device is provided. The information processing system includes a feature value acquisition unit, a spectral decomposition unit, and a synchronization processing unit. The feature value acquisition unit is configured to acquire waveform data, the waveform data being indicated as feature values obtained from each of the plurality of projection images. The spectral decomposition unit is configured to acquire decomposed waveform data by applying a predetermined analytical method to the waveform data, the decomposed waveform data being data of a waveform obtained by decomposing the waveform data. The synchronization processing unit is configured to determine a projection image to be used for reconstruction from among the plurality of projection images based on the waveform indicated by the decomposed waveform data.