Machine Learning-Assisted Raman Spectroscopy for pH and Lactate Sensing in Body Fluids

Abstract

This study presents the combination of Raman spectroscopy with machine learning algorithms as a prospective diagnostic tool capable of detecting and monitoring relevant variations of pH and lactate as recognized biomarkers of several pathologies. The applicability of the method proposed here is tested both in vitro and ex vivo. In a first step, Raman spectra of aqueous solutions are evaluated for the identification of characteristic patterns resulting from changes in pH or in the concentration of lactate. The method is further validated with blood and plasma samples. Principal component analysis is used to highlight the relevant features that differentiate the Raman spectra regarding their pH and concentration of lactate. Partial least squares regression models are developed to capture and model the spectral variability of the Raman spectra. The performance of these predictive regression models is demonstrated by clinically accurate predictions of pH and lactate from unknown samples in the physiologically relevant range. These results prove the potential of our method to develop a noninvasive technology, based on Raman spectroscopy, for continuous monitoring of pH and lactate in vivo.

Identifying the charge density and dielectric environment of graphene using Raman spectroscopy and deep learning

“…They are capable of efficiently handling large amounts of data and detecting complex nonlinear relationships 17 . With the development of deep learning, more deep neural networks have been proposed and applied to Raman spectral analysis [18][19][20][21][22][23] .…”

Design and Development of a Bimodal Optical Instrument for Simultaneous Vibrational Spectroscopy Measurements

“…In the field of medical diagnostics and biomedical topics, vibrational spectroscopy has proven its value in combination with multivariate analysis and chemometrics and is now stepwise entering into clinical applications [ 12 ]. The ability of detecting biochemical changes in cells and tissues allows for real-time disease diagnostics.…”

Experimental characterization, machine learning analysis and computational modelling of the high effective inhibition of copper corrosion by 5‐(4‐pyridyl)‐1,3,4‐oxadiazole‐2‐thiol in saline environment

“…In addition, a machine learning method based on the Raman data obtained by analyzing the copper surface after its immersion at different times in the blank and PyODT-containing solutions was applied to follow the evolution of the inhibitor film adsorbed on the metal, in correlation with the inhibitor concentration and the exposure time. The combination of Raman spectroscopy data and Machine Learning algorithms has already been reported [40,41] as a prospective tool capable of identifying adulterations in foods and beverages and for biological detection, in forensic and medical diagnostics as well as materials analysis, because of its power to detect patterns in complex data sets. However, this is the first time that Machine Learning algorithms have been used in corrosion studies to distinguish between Raman profiles of corroded and inhibitor-protected metal samples.…”