Ultrasound Preset Parameter Learning for Consistent Image Quality

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Ultrasound image quality is highly operator-dependent due to numerous adjustable parameters, and inexperienced users struggle to achieve desired image quality using preset parameter values, leading to inefficiencies and potential misdiagnosis.

Innovation Solution

A federated learning approach is employed to generate and update preset parameter values for ultrasound imaging systems by combining local machine-learning models from multiple systems into a global model, incorporating operator feedback to standardize and personalize image quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual adjustment of parameters is used to achieve desired image quality, then image quality can be optimized, but time consumption and operator expertise requirements increase

Engineering Contradiction:

Improveimage qualityVSAvoidtime to adjust parameters

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system pre-calculates and stores optimal parameter sets for different imaging scenarios before actual use. When imaging is required, the pre-computed parameters are directly applied, eliminating the need for real-time manual adjustment and reducing time consumption while maintaining image quality optimization

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system automatically selects and adjusts parameters based on the imaging task and environmental conditions without requiring operator intervention. The autonomous parameter selection mechanism serves itself by learning from operator feedback and continuously improving its parameter recommendations, thereby reducing both time consumption and expertise requirements

Inventive Principle:

Principle #25Self-service

2Manufacturing precision

If manual adjustment of parameters is used to achieve desired image quality, then image quality can be optimized, but operator expertise requirements increase

Engineering Contradiction:

Improveimage qualityVSAvoidoperator expertise requirement

Core Design Contradiction:

Manufacturing precisionVSEase of operation

Solution Approach 1:

The system autonomously selects optimal parameters based on imaging conditions and task type without requiring operator knowledge of parameter tuning. The self-service mechanism includes automatic parameter selection, adaptive adjustment based on feedback, and intelligent recommendations that eliminate the need for specialized operator expertise while maintaining high image quality

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces manual mechanical parameter adjustment with automated electronic control and machine learning algorithms. This substitution transforms the manual operation into an automated process that intelligently determines optimal parameters, thereby reducing operator expertise requirements while maintaining image quality optimization

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If fixed preset parameter values are used, then ease of operation improves, but adaptability to different conditions deteriorates

Engineering Contradiction:

Improveease of useVSAvoidadaptability to conditions

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system transforms static fixed presets into dynamic adaptive parameter sets that automatically adjust based on imaging conditions, task type, and environmental factors. The parameters are no longer fixed but dynamically optimized through machine learning models that learn from operator feedback and continuously adapt to different scenarios, thereby maintaining ease of operation while improving adaptability

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms where operator adjustments and imaging outcomes are used to continuously refine and update the parameter recommendations. This feedback loop enables the system to adapt to different conditions by learning from actual usage patterns and performance data, maintaining both ease of operation and adaptability simultaneously

Inventive Principle:

Principle #23Feedback

4Adaptability or versatility

If local machine-learning models are trained independently at each system, then adaptability to local conditions improves, but manufacturing precision deteriorates due to lack of standardization

Engineering Contradiction:

Improvelocal adaptabilityVSAvoidimage quality consistency

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The system merges multiple local machine-learning models into a unified global model that consolidates knowledge from different imaging environments and operators. This combined model maintains the adaptability benefits of local training while achieving standardized performance across all systems through shared learning, thereby resolving the contradiction between local adaptability and global consistency

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS12611188B2Defining preset parameter values for an ultrasound imaging system

Publication Date: 2026.04.28 KONINKLIJKE PHILIPS NV

AI summary

A mechanism for defining a set of preset parameter values for an ultrasound imaging system. Information about local machine-learning models, generated by a plurality of ultrasound imaging systems and updated responsive to operator feedback, is provided to an external server. The external server generates a global machine-learning model based on this information, which is then used to update the local machine-learning model on target ultrasound imaging systems.