Amplification Curve Pattern Matching for Noisy Molecular Diagnostics

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing molecular diagnostic methods using nucleic acid amplification face challenges in accurately estimating Ct values, quantitation results, and positive/negative determinations due to interference from background signals and noise, especially when multiple target nucleic acids are present, leading to inefficiencies in time and cost.

Innovation Solution

A method utilizing a computer device with a pre-trained estimation model that calculates shape similarity between amplification curves and reference patterns, employing neural networks to estimate molecular diagnostic analysis results, including Ct, quantitative values, and suitability of oligonucleotide candidates, thereby reducing the need for conventional processes like standard curve experiments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If conventional methods (Ct calculation, standard curve experiments) are used for molecular diagnostic analysis, then diagnostic results can be obtained, but the process is time-consuming and costly due to multiple experimental steps and standard curve requirements

Engineering Contradiction:
Improvetime for diagnostic analysisVSAvoidaccuracy of diagnostic results
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent performs preliminary actions by training the neural network model in advance using datasets containing amplification curves and corresponding diagnostic results. This pre-trained model can then directly estimate Ct values, quantitation results, and positive/negative determinations from new amplification curves without requiring time-consuming standard curve experiments or multiple processing steps, thus reducing analysis time while maintaining diagnostic accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a digital copy or model of the diagnostic analysis process through the neural network. Instead of performing physical standard curve experiments, the system uses the trained neural network model to replicate and predict diagnostic results from amplification curves, eliminating the need for repeated experimental procedures while preserving measurement precision

Inventive Principle:
Principle #26Copying

2Measurement precision

If standard curve experiments are performed for accurate quantitation, then quantitative results can be obtained, but the process becomes more complex and time-consuming

Engineering Contradiction:
Improvequantitation accuracyVSAvoidcomplexity of experimental process
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and removes the need for standard curve experiments from the diagnostic process. The neural network model has been trained to directly estimate quantitation results from amplification curves alone, extracting the essential information needed for accurate quantitation without requiring the separate, complex standard curve preparation and execution steps

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the approach from physical experimental parameters (standard curve concentrations, serial dilutions) to computational parameters (neural network weights, activation functions). The model learns the relationship between amplification curve characteristics and quantitation results through training, transforming the quantitation process from a physical experiment to a computational estimation that maintains accuracy while reducing complexity

Inventive Principle:
Principle #35Parameter changes

3Reliability

If multiple processing steps are used for molecular diagnostic analysis, then comprehensive results can be obtained, but the overall process efficiency decreases

Engineering Contradiction:
Improvecomprehensiveness of diagnostic resultsVSAvoidefficiency of diagnostic analysis
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent merges multiple separate diagnostic functions (Ct value calculation, quantitation estimation, positive/negative determination) into a single integrated neural network model. The model simultaneously performs all these tasks by processing the amplification curve through one unified computational framework, eliminating the need for sequential processing steps and thereby improving efficiency while maintaining comprehensive diagnostic reliability

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4648054A1Method for obtaining molecular diagnostic analysis results, method for obtaining model to estimate molecular diagnostic analysis results, and computer device for performing same
Publication Date: 2025.11.12 SEEGENE INC
  • EP4648054A1 patent drawingFigure 1
  • EP4648054A1 patent drawingFigure 2
  • EP4648054A1 patent drawingFigure 3

AI summary

Provided according to one embodiment is a method for acquiring molecular diagnostic analysis results, performed by a computer device using a memory, a processor, and one or more programs stored in the memory and configured to be executed by the processor, the method comprising the steps of: acquiring a dataset representing results of the amplification reaction for a target analyte in the sample; calculating the shape similarity for each reference pattern by comparing the target curve in the dataset to multiple pre-established reference patterns; and providing the shape similarity for each reference pattern to a pre-trained estimation model, and acquiring, from the estimation model, molecular diagnostic analysis results including at least one of the Ct of the target curve, the quantitative value of the target analyte in the sample, the positive/negative reading result for the target analyte in the sample, and the suitability assessment result of oligonucleotide candidates used in the amplification reaction.