Thyroid Nodule Classifier Algorithm for Malignancy Diagnosis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for distinguishing benign and malignant thyroid nodules are not entirely effective, leading to unnecessary surgeries, as they often classify nodules as indeterminate, and existing gene-based classifiers lack sufficient validation in European populations.

Innovation Solution

A method using a classifier algorithm that evaluates the expression levels of specific genes (DCSTAMP, NOD1, SLC26A7, EMP2, EGR1, MPZL2, ITGA2, MET, SLPI, and TPO) in thyroid nodule samples, combined with clinical features, to accurately differentiate between benign and malignant nodules with high accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If fine-needle aspiration biopsy (FNAB) is used for diagnosis, then the diagnostic procedure is minimally invasive and easy to perform, but the ability to assess follicular cell arrangement and invasive features is lost, leading to indeterminate results

Engineering Contradiction:
Improveease of biopsy procedureVSAvoiddiagnostic accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces molecular markers (gene expressions, mutations, and microRNA profiles) as intermediary indicators that bridge the gap between minimally invasive FNAB sampling and accurate malignancy assessment. These molecular markers serve as mediators that provide additional diagnostic information without requiring more invasive procedures, thus maintaining ease of operation while improving measurement precision.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/cytological assessment limitations of FNAB with molecular biological analysis. Instead of relying solely on visual evaluation of cell morphology and arrangement (mechanical observation), the invention uses molecular markers (gene expressions, mutations, microRNA profiles) to substitute and enhance the diagnostic capability, enabling accurate distinction between benign and malignant nodules even with limited cellular material.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If molecular markers are used to improve diagnostic accuracy, then the ability to distinguish benign from malignant nodules improves, but the complexity of the diagnostic procedure increases

Engineering Contradiction:
Improvediagnostic accuracyVSAvoiddiagnostic procedure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the molecular diagnostic panel into distinct, manageable components: gene expression analysis (e.g., HNF1A, PAX8, TTF1), mutation detection (e.g., BRAF V600E, RAS mutations), and microRNA profiling. This segmentation allows the complex diagnostic information to be processed in modular fashion, making the overall procedure more manageable and interpretable while maintaining high diagnostic accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent develops a universal molecular marker panel that can be applied across different thyroid nodule cases regardless of cytological category (B3, B4, or B5). This multi-functional approach allows the same set of molecular markers to provide diagnostic value across various clinical scenarios, reducing the need for multiple different test protocols and thereby managing complexity while improving accuracy.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If a larger number of molecular markers are analyzed, then the diagnostic accuracy increases, but the cost and time required for the test increase

Engineering Contradiction:
Improvediagnostic accuracyVSAvoiddiagnostic time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary analysis by first categorizing the nodule based on cytological features (Bethesda system classification), then selectively applying molecular marker analysis based on the pre-test probability of malignancy. This preliminary stratification allows the system to focus molecular testing on cases where it will provide the most value, reducing overall diagnostic time while maintaining high accuracy for indeterminate cases.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent optimizes the molecular marker panel by selecting a specific set of markers with high discriminatory power (e.g., 5-10 key genes, specific mutations, and microRNAs) rather than analyzing all possible molecular markers. This parameter optimization balances diagnostic accuracy with testing time and cost, identifying the minimum effective set of markers needed for reliable classification.

Inventive Principle:
Principle #35Parameter changes

4Loss of information

If existing gene classifiers are used, then some diagnostic information is obtained, but their performance is insufficient in European populations due to lack of validation

Engineering Contradiction:
Improvediagnostic informationVSAvoidpopulation-specific reliability
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The patent creates a dynamic, adaptable molecular marker panel that can be adjusted and validated for different population characteristics. Rather than using a fixed classifier developed in one population, the invention allows for optimization of marker selection and threshold values based on European population data, ensuring the diagnostic system remains reliable and accurate across different genetic backgrounds and epidemiological contexts.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4303324A1A method of distinguishing between benign and malignant thyroid nodules
Publication Date: 2024.01.10 WASKO SPOLKA AKCYJNA (WASKO SA)
  • EP4303324A1 patent drawingFigure 1
  • EP4303324A1 patent drawingFigure 2
  • EP4303324A1 patent drawingFigure 3

AI summary

A method of distinguishing benign and malignant thyroid nodules in the nodule biopsy sample, classified according to the Bethesda system, including the steps of RNA isolation and the determination of expression level of a normalized gene set, including AMOT, DCSTAMP, NOD1, SLC26A7, EMP2, EGR1, MPZL2, ITGA2, MET, SLPI, TPO. To differentiate thyroid nodules, a classifier algorithm is used, which is trained on gene expression data correlated with the nature of the thyroid nodule. In addition, the probability of malignancy of the test sample is determined.