Biomarker Intensity Prediction From Stain Signals With Confidence

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing digital pathology techniques struggle to quantify biomarker signals accurately due to high cell density, aggregation, and saturation issues, leading to unreliable diagnoses and treatment recommendations.

Innovation Solution

A computer-implemented method using a linear and non-linear confidence function to predict biomarker intensities from stain intensities, generating a confidence metric, and outputting results based on these predictions, along with error metrics, to establish an artificial ground truth for quantifying biomarker levels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If digital pathology images are used to quantify biomarker signals, then quantitative information can be obtained, but measurement precision deteriorates due to cell density, aggregation, and saturation issues

Engineering Contradiction:

Improvebiomarker signal quantification accuracyVSAvoiddiagnosis and treatment recommendation reliability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent transforms the measurement approach by changing from direct signal intensity measurement to a two-step process: first measuring stain intensity, then using a pre-generated prediction function to calculate biomarker intensity. This parameter transformation resolves the saturation and aggregation problems by decoupling the measurement process from the biological signal saturation issues.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces a stain intensity as an intermediary measurement that indirectly reflects biomarker intensity. Instead of directly measuring the biomarker signal which suffers from saturation and aggregation, the system measures the stain intensity and uses a prediction function to infer the biomarker intensity, thereby avoiding the direct measurement limitations.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If relative or qualitative signal assessment is used, then analysis can be performed, but quantitative information for diagnosis and treatment is lost

Engineering Contradiction:

Improveanalysis throughputVSAvoidquantitative biomarker measurement accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent replaces manual qualitative assessment with an automated computational system that performs quantitative analysis. The image analysis system automatically measures stain intensity, applies prediction functions, and generates quantitative biomarker intensity values, substituting the mechanical manual process with an automated digital system that maintains both throughput and precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Quantity of substance

If high cell density or aggregation is present, then biological signal is strong, but signal quantification becomes unreliable due to saturation

Engineering Contradiction:

Improvebiomarker expression levelVSAvoidsignal intensity measurement accuracy

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent uses stain intensity as an intermediary that linearly relates to biomarker intensity through a prediction function. This intermediary approach allows the system to accurately quantify biomarkers even when direct signal measurement would be saturated, as the stain intensity measurement remains within a linear, measurable range.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the measurement parameter from direct biomarker signal intensity to stain intensity. This parameter transformation allows the system to accurately represent high biomarker expression levels without suffering from saturation, as the stain intensity measurement scale can accommodate the full range of expression levels linearly.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20260011013A1Ground truth of signal aggregate quantification

Publication Date: 2026.01.08 VENTANA MEDICAL SYSTEMS INC

AI summary

A digital pathology image collected using bright-field imaging that depicts a slide with a stained sample slice is accessed. A stain intensity that corresponds to at least part of the digital pathology image is detected. A biomarker-intensity-prediction function that linearly relates predicted biomarker-intensity levels to detected stain intensities is accessed. A non-linear confidence function is accessed that relates confidences of a predicted biomarker intensity to the detected intensities of the stain. A predicted biomarker intensity is generated for the at least part of the slide using the detected stain intensity that corresponds to the at least part of the slide and the linear biomarker-intensity-prediction function. A confidence metric for the predicted biomarker intensity is generated using the detected stain intensity that corresponds to the at least part of the slide and based on the confidence function. A result based on the predicted biomarker intensity and the confidence metric is output.