Gaussian-Lorentzian fitting

Table of contents

  1. Introduction
  2. How to use
  3. Behavior
  4. Method
  5. References

Introduction

Gaussian-Lorentzian fitting estimates a baseline from user-selected fitting ranges using a mixed peak-shape model. It is useful when the baseline can be approximated from selected spectral regions.

How to use

  1. Upload data and open Processing Page.
  2. Enable Baseline removal.
  3. In Select baseline removal function, choose Gaussian-Lorentzian fitting.
  4. Set Number of fitting ranges.
  5. Enter the Start of range and End of range values.
  6. Click Apply fitting ranges, then process the data.

Behavior

SpectraGuru validates and clips fitting ranges to the data bounds, warns about invalid or overlapping ranges, fits the selected ranges, and subtracts the estimated baseline from each selected spectrum.

Method

The fitted baseline uses a mixed Gaussian-Lorentzian shape:

\[b(x)=G(x;A,\mu,\sigma)+L(x;S,\mu,\gamma)\]
Parameter Tunable or fixed Implementation
Number of fitting ranges Tunable Default 2, UI range 2-10
Start/end range bounds Tunable Clipped to dataset Raman shift bounds and checked for overlap
Curve fitting Fixed Nonlinear least-squares fitting with user ranges
Baseline action Fixed Fitted baseline is subtracted from each selected spectrum

References

  1. Chen, H., et al. (2022). Rapid and quantitative detection of respiratory viruses using surface-enhanced Raman spectroscopy and machine learning. Biosensors and Bioelectronics, 202, 114721. https://doi.org/10.1016/j.bios.2022.114721
  2. SciPy Developers. scipy.optimize.curve_fit. https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.curve_fit.html