Mammography Imaging Conditions Based on Breast Composition Index

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing mammography apparatuses rely solely on direct information from pre-imaging for determining imaging conditions, which may not account for variations in breast composition, leading to suboptimal conditions for breasts of similar thickness.

Innovation Solution

A mammography apparatus that acquires additional information beyond pre-imaging, including breast state and subject information, to calculate an index for determining more appropriate imaging and image processing conditions using a trained model.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If only direct information from pre-imaging is used for determining imaging conditions, then the determination process is simple and quick, but the accuracy of imaging condition determination deteriorates because variations in breast composition are not accounted for

Engineering Contradiction:

Improveaccuracy of imaging condition determinationVSAvoidcomplexity of information processing

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs pre-imaging first to obtain preliminary information about breast thickness and basic characteristics. This preliminary action allows the system to prepare and select appropriate deep learning models in advance, which are then used to analyze the pre-image and extract additional breast composition information before final imaging condition determination.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

A deep learning model acts as an intermediary between the pre-imaging data and the imaging condition determination. The model processes the pre-image to extract additional information about breast composition (fatty vs. dense breast), which then informs the selection of optimal imaging conditions. This intermediary enables accurate determination without requiring complex manual analysis.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Manufacturing precision

If imaging conditions are standardized for breasts of the same thickness, then the operation is simple, but the image quality deteriorates because different breast compositions require different imaging conditions

Engineering Contradiction:

Improveimage qualityVSAvoidsimplicity of condition selection

Core Design Contradiction:

Manufacturing precisionVSEase of operation

Solution Approach 1:

The system applies different imaging conditions to different local characteristics of the breast. By analyzing the pre-image with deep learning models, the system identifies whether the breast is fatty or dense in specific regions and selects imaging conditions tailored to each composition type, rather than applying a uniform standard to all breasts of the same thickness.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes imaging parameters (such as X-ray energy level, compression force, and exposure settings) based on the detected breast composition. For fatty breasts, different parameters are applied compared to dense breasts, allowing optimization of image quality for each specific case while the deep learning model automates the parameter selection process.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If additional information beyond pre-imaging is acquired to determine imaging conditions, then the accuracy of condition determination improves, but the examination time increases

Engineering Contradiction:

Improveaccuracy of breast composition analysisVSAvoidexamination time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs the additional analysis during the pre-imaging phase, which is conducted anyway as part of the standard examination protocol. The deep learning model processes the pre-image data that is already captured, extracting additional breast composition information without requiring separate imaging sessions or additional patient time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The deep learning analysis continues seamlessly after pre-imaging by automatically processing the captured pre-image data. This continuous processing extracts additional breast composition information and feeds it into the imaging condition determination without interrupting the examination flow or requiring additional patient presence time.

Inventive Principle:

Principle #20Continuity of useful action

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

Enables accurate determination of imaging conditions tailored to breast density and composition, reducing radiation dose variation and enhancing image quality, allowing for more efficient and accurate mammography procedures.

Implementation Method 1

the pixel value of an image acquired by the pre-imaging (hereinafter, a 'pre-image')... the thickness of the breast under compression... breast density from a main image

Methodology Applied

Scientific EffectX-ray absorption: Absorption (EM radiation)

Data Source

PatentUS12582369B2Mammography apparatus

Publication Date: 2026.03.24 CANON MEDICAL SYST CORP

AI summary

According to one embodiment, a mammography apparatus includes processing circuitry. The processing circuitry is configured to acquire, as information on preliminary X-ray imaging for a breast of a subject, first information including at least one of an imaging condition, breast state information indicating a state of the breast, and/or information on the subject. The processing circuitry is configured to calculate an index for the breast based on the first information.