Image Measurement Endpoint Prediction for Consistent Medical Sizing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods for making measurements in images, particularly medical images, are costly, subjective, and suffer from inter-reader variability due to complex and subjective measurement criteria that are difficult to algorithmically capture, necessitating improved automated methods.

Innovation Solution

A computer-implemented method using a machine learning model trained on annotated images to predict appropriate endpoints for measurements, combining the expertise of multiple annotators and reducing subjectivity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional interactive caliper measurements are used by radiologists, then measurement flexibility and adaptability to complex criteria are maintained, but workload burden increases and inter-reader variability persists

Engineering Contradiction:

Improvemeasurement flexibilityVSAvoidworkload burden

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The system enables self-service automation where the measurement tool automatically identifies landmarks and performs measurements based on trained machine learning models, eliminating the need for manual caliper placement while maintaining adaptability to complex measurement criteria through the trained model's understanding of radiological conventions

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical interaction of radiologists placing calipers with an automated computer vision system that uses machine learning models to detect landmarks and calculate measurements, substituting human manual operations with automated image processing algorithms

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If conventional interactive caliper measurements are used by radiologists, then adaptability to complex measurement criteria is maintained, but subjectivity and inter-reader variability increase

Engineering Contradiction:

Improveadaptability to measurement criteriaVSAvoidinter-reader variability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system transforms complex, subjective measurement criteria into quantifiable parameters that machine learning models can process, converting radiological conventions into numerical features and decision rules that the model learns from annotated training data, thereby standardizing the interpretation of complex criteria

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system creates a digital copy of expert radiologist knowledge by training machine learning models on annotated measurements from multiple radiologists, capturing their expertise and decision-making patterns in a reusable model that consistently applies the same measurement logic across all images

Inventive Principle:

Principle #26Copying

3Productivity

If automated measurement methods are implemented, then productivity and consistency are improved, but ability to handle complex caveated measurement criteria deteriorates

Engineering Contradiction:

Improvemeasurement efficiencyVSAvoidhandling complex criteria

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary action by pre-training machine learning models on extensive annotated datasets that encode complex measurement criteria and caveats before deployment, allowing the automated system to handle complex criteria efficiently during actual measurement tasks without requiring real-time expert intervention

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces machine learning models as intermediaries between the raw image data and the measurement output, where the models learn to interpret complex measurement criteria and translate them into automated measurement decisions, bridging the gap between simple automation and complex radiological conventions

Inventive Principle:

Principle #24Intermediary (Mediator)

4Adaptability or versatility

If manual caliper placement by radiologists is used, then adaptability to specific measurement needs is maintained, but measurement precision and accuracy deteriorate due to subjectivity

Engineering Contradiction:

Improvemeasurement adaptabilityVSAvoidmeasurement accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system creates a digital copy of expert measurement judgment by training machine learning models on annotated data from multiple radiologists, capturing their expertise in selecting appropriate measurement endpoints and applying measurement conventions, thereby achieving consistent high-precision measurements that replicate expert accuracy

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system transforms subjective measurement decisions into objective parameter-based decisions by training models to recognize visual features and patterns that correlate with correct measurement endpoints, converting radiologist intuition into quantifiable image features and decision boundaries

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentEP4078611B1Making measurements in images

Publication Date: 2026.02.11 KONINKLIJKE PHILIPS NV

Figure 1(a)~1(b)
Figure 1(c)
Figure 2(a)~2(b)

AI summary

A computer implemented method of making a measurement associated with a feature of interest in an image. The method comprises using (302) a model trained using a machine learning process to take the image as input and predict a pair of points between which to make the measurement of the feature of interest in the image. The method then comprises determining (304) the measurement, based on the predicted pair of points.