Medical Image Consistency Checking Using Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Inconsistencies between imaging order information and obtained medical images can impair the workflow in medical image processing systems, often noticed only after the patient has left, necessitating re-imaging and inefficiencies.

Innovation Solution

A medical image processing apparatus using machine learning to determine consistency between imaging order information and obtained medical images, including an inference unit for accurate inference and a verification unit to update parameters, ensuring correct image capture and storage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual checking of imaging order information and medical images is performed, then workflow simplicity is maintained, but inconsistency detection accuracy deteriorates

Engineering Contradiction:

Improveinconsistency detection accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual visual inspection and comparison of imaging order information with automated machine learning-based image recognition. The system extracts anatomical structures from medical images and compares them with expected structures from imaging orders, substituting human judgment with algorithmic analysis to achieve consistent and accurate inconsistency detection.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If re-imaging is performed to correct inconsistencies, then image quality reliability is improved, but productivity deteriorates

Engineering Contradiction:

Improveimage capture accuracyVSAvoidworkflow efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system performs inconsistency detection immediately after image acquisition using automated machine learning analysis, identifying errors while the patient is still present. This preliminary detection prevents delayed discovery of inconsistencies and eliminates the need for time-consuming re-imaging procedures, thereby maintaining both reliability and productivity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a feedback mechanism where the machine learning system analyzes acquired images, compares them with imaging order requirements, and provides immediate feedback on consistency. This closed-loop feedback enables real-time quality control without interrupting the workflow, ensuring that only consistent images proceed to storage and viewing.

Inventive Principle:

Principle #23Feedback

3Loss of information

If information is updated in one location (image server or modality), then data currency is improved, but information consistency deteriorates

Engineering Contradiction:

Improveinformation currencyVSAvoidinformation consistency

Core Design Contradiction:

Loss of informationVSStability of the object's composition

Solution Approach 1:

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12548148B2Medical image processing apparatus, medical image processing method, and storage medium

Publication Date: 2026.02.10 CANON KK

AI summary

A medical image processing apparatus includes an obtaining unit configured to obtain a medical image based on imaging order information, and a determination unit configured to determine, using parameters obtained by machine learning, consistency between the imaging order information and the medical image.