NMR Signal Processing Optimizing Peak Intensity Estimation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing NMR signal processing methods face challenges in accurately determining the intensity of peaks in spectra due to issues with integral range determination, initial value dependence, and estimation accuracy, particularly when dealing with overlapping peaks and noise components.

Innovation Solution

A system and method that optimize the estimation order of a mathematical model representing an FID signal by varying the number of signal components, evaluating results, and selecting optimal parameters to accurately estimate peak intensities, thereby improving estimation accuracy and reducing errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the integration method is used to find peak intensity by setting a frequency range, then the intensity can be calculated, but it is difficult to determine the integral range and the computed intensity varies depending on the integral range setting

Engineering Contradiction:

Improvepeak intensity measurement accuracyVSAvoidintegral range determination difficulty

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent replaces the manual integration method with automated waveform separation using curve fitting. Instead of requiring user-defined integral ranges, the system uses mathematical models (Lorentzian, Gaussian, or Voigt functions) to automatically separate and integrate overlapping peaks, eliminating the subjectivity and difficulty of manual range selection while improving measurement precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If waveform separation method (curve fitting) is used to approximate peaks, then peak intensity can be found, but computational results depend on the initial value of parameters such as intensity, linewidth, or frequency

Engineering Contradiction:

Improvepeak intensity measurement accuracyVSAvoiddependence on initial parameter values

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent applies preliminary signal processing steps including phase correction and baseline correction before curve fitting. These preliminary actions prepare the data in a standardized format that reduces sensitivity to initial parameter values. Additionally, the system uses automated parameter estimation routines that provide more reliable starting points for the optimization algorithm, thereby reducing dependence on user-defined initial values.

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If the Prony method is used to estimate signal components, then it avoids integral range setting problems, but estimation results depend on the number of assumed signal components (order)

Engineering Contradiction:

Improveautomatic parameter estimationVSAvoidestimation accuracy dependence on order

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent implements a dynamic order selection mechanism that automatically determines the optimal number of signal components. The system evaluates multiple possible orders and selects the one that best fits the data using statistical criteria, transforming the static Prony method into an adaptive system that dynamically adjusts the model complexity based on the actual signal characteristics, thereby resolving the contradiction between automation and accuracy.

Inventive Principle:

Principle #15Dynamics

4Measurement precision

If high-order estimation method is used by assuming more signal components than peaks of interest, then it avoids integral range setting problems and initial value dependence, but estimation results still depend on the number of assumed signal components

Engineering Contradiction:

Improveavoidance of integral range and initial value problemsVSAvoidcomplexity of determining optimal order

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent incorporates feedback mechanisms where the system evaluates the quality of estimation results and adjusts the number of assumed signal components accordingly. By monitoring fit quality metrics and using statistical criteria, the system provides feedback to refine the model order, automatically converging to the optimal complexity without requiring complex manual intervention, thus resolving the contradiction between avoiding manual parameter setting and determining optimal model complexity.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS10451695B2System and method for processing NMR signals

Publication Date: 2019.10.22 JEOL LTD

AI summary

There is disclosed an NMR signal processing method for accurately estimating the intensities of p peaks of interest in an NMR spectrum by the use of a mathematical model that represents a time-domain, free induction decay (FID) signal obtained by an NMR measurement as a sum of q signal components. First, q parameters (each being a combination of a pole and a complex intensity) defining q signal components are estimated for each value of the estimation order q of the mathematical model while varying the value of the estimation order q (S34). At each value of the estimation order q, p parameters are selected from the q parameters in accordance with selection criteria (S42, S46). The selected p parameters are evaluated (S48). An optimal value of the estimation order is determined based on the evaluation values produced at the various values of the estimation order q, and p parameters corresponding to the optimal value of the estimation order is identified. The intensities of p peaks of interest are found from the identified p parameters.