X-ray CT Inference Selection for Diagnostic Consistency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The inference results from X-ray computed tomography (CT) apparatuses equipped with a photon counting CT function vary depending on whether material decomposition image data or CT image data is used, leading to inconsistent diagnostic outputs.

Innovation Solution

An X-ray CT apparatus with multiple inference units performing tasks on material decomposition image data, CT image data, and composite image data, along with a selection unit that chooses the appropriate inference result for display based on differences between these results, ensuring accurate and consistent diagnostic outputs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple inference units process different image data types (material decomposition image data, CT image data, composite image data), then the accuracy and appropriateness of inference results are improved, but the device complexity increases due to the need for multiple inference units and a selection unit

Engineering Contradiction:
Improveinference result accuracyVSAvoidsystem structure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system divides the inference processing into three separate inference units, each dedicated to processing a specific type of image data (material decomposition image data, CT image data, and composite image data). This segmentation allows each unit to specialize in its data type, improving inference accuracy while maintaining manageable complexity through functional separation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A selection unit is introduced as an intermediary component that receives inference results from all three inference units, compares them, and selects the most appropriate result for output. This mediator resolves the complexity of directly integrating multiple inference outputs by providing a centralized decision-making mechanism.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Stability of the object's composition

If inference is performed on multiple types of image data, then diagnostic consistency is improved, but the processing time and loss of time increase due to multiple inference operations

Engineering Contradiction:
Improvediagnostic consistencyVSAvoidprocessing time
Core Design Contradiction:
Stability of the object's compositionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by simultaneously executing all three inference operations on their respective image data types before the selection unit compares results. This parallel preliminary processing ensures that all necessary inference work is completed upfront, maintaining diagnostic consistency while minimizing subsequent processing time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system merges the results from three separate inference operations into a single selected output through the selection unit. By combining multiple inference outcomes and selecting the best one, the system achieves diagnostic consistency without requiring sequential processing of all inference types.

Inventive Principle:
Principle #5Merging (Combining)

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

The solution ensures the display of appropriate inference results by selecting the most accurate outcome from material decomposition, CT, or composite image data inference, enhancing diagnostic consistency and accuracy.

Implementation Method 1

an X-ray tube configured to radiate X-rays; a detector configured to detect X-rays radiated by the X-ray tube and passed through a subject

Methodology Applied
Scientific EffectX-ray detection: X-Ray

Data Source

PatentUS20240148344A1X-ray computed tomography apparatus, information processing system, information processing method, and non-transitory computer-readable medium
Publication Date: 2024.05.09 CANON KK
  • US20240148344A1 patent drawing
  • US20240148344A1 patent drawing
  • US20240148344A1 patent drawing

AI summary

An X-ray computed tomography (CT) apparatus includes an X-ray tube configured to radiate X-rays; a detector configured to detect X-rays radiated by the X-ray tube and passed through a subject; a reconstruction processing unit configured to reconstruct imaging data output by the detector and generate CT image data and material decomposition image data; a first inference unit configured to perform inference on the material decomposition image data; a second inference unit configured to perform inference on the CT image data; a third inference unit configured to perform inference on composite image data; a selection unit configured to, in a case where an inference result of the first inference unit is different from an inference result of the second inference unit, select an inference result of the third inference unit; and a display control unit configured to cause the selected inference result to be displayed.