Premium
A representation learning approach for recovering scatter‐corrected spectra from Fourier‐transform infrared spectra of tissue samples
Author(s) -
Raulf Arne P.,
Butke Joshua,
Menzen Lukas,
Küpper Claus,
Großerueschkamp Frederik,
Gerwert Klaus,
Mosig Axel
Publication year - 2021
Publication title -
journal of biophotonics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.877
H-Index - 66
eISSN - 1864-0648
pISSN - 1864-063X
DOI - 10.1002/jbio.202000385
Subject(s) - fourier transform , scattering , preprocessor , spectral line , infrared , fourier transform infrared spectroscopy , biological system , computation , infrared spectroscopy , computer science , fast fourier transform , artificial intelligence , materials science , physics , pattern recognition (psychology) , optics , computational physics , algorithm , mathematics , mathematical analysis , biology , astronomy , quantum mechanics
Infrared spectra obtained from cell or tissue specimen have commonly been observed to involve a significant degree of scattering effects, often Mie scattering, which probably overshadows biochemically relevant spectral information by a nonlinear, nonadditive spectral component in Fourier transform infrared (FTIR) spectroscopic measurements. Correspondingly, many successful machine learning approaches for FTIR spectra have relied on preprocessing procedures that computationally remove the scattering components from an infrared spectrum. We propose an approach to approximate this complex preprocessing function using deep neural networks. As we demonstrate, the resulting model is not just several orders of magnitudes faster, which is important for real‐time clinical applications, but also generalizes strongly across different tissue types. Using Bayesian machine learning approaches, our approach unveils model uncertainty that coincides with a band shift in the amide I region that occurs when scattering is removed computationally based on an established physical model. Furthermore, our proposed method overcomes the trade‐off between computation time and the corrected spectrum being biased towards an artificial reference spectrum.