Optimisation of wavelength modulated Raman spectroscopy: towards high throughput cell screening.

Bavishna B Praveen,Kishan Dholakia,Michael Mazilu,Robert F Marchington,C Simon Herrington,Andrew Riches,Arrate Muñoz-Barrutia

doi:10.1371/journal.pone.0067211

Bavishna B Praveen, Kishan Dholakia + Show 5 more

Open Access

https://doi.org/10.1371/journal.pone.0067211

Copy DOI

Abstract

In the field of biomedicine, Raman spectroscopy is a powerful technique to discriminate between normal and cancerous cells. However the strong background signal from the sample and the instrumentation affects the efficiency of this discrimination technique. Wavelength Modulated Raman spectroscopy (WMRS) may suppress the background from the Raman spectra. In this study we demonstrate a systematic approach for optimizing the various parameters of WMRS to achieve a reduction in the acquisition time for potential applications such as higher throughput cell screening. The Signal to Noise Ratio (SNR) of the Raman bands depends on the modulation amplitude, time constant and total acquisition time. It was observed that the sampling rate does not influence the signal to noise ratio of the Raman bands if three or more wavelengths are sampled. With these optimised WMRS parameters, we increased the throughput in the binary classification of normal human urothelial cells and bladder cancer cells by reducing the total acquisition time to 6 s which is significantly lower in comparison to previous acquisition times required for the discrimination between similar cell types.

Highlights

Raman spectroscopy is a vibrational spectroscopic technique that provides information regarding the chemical composition of a sample of interest [1]
With a modulation amplitude exceeding Dl = 0.32 nm (Dn = 160 GHz), the Raman bands of biological samples are resolvable
It was observed that the Signal to Noise Ratio (SNR) of Raman bands increases with the time constant and the total acquisition time

Summary

Introduction

Raman spectroscopy is a vibrational spectroscopic technique that provides information regarding the chemical composition of a sample of interest [1]. The weak Raman signal is obscured by the luminescence background resulting from the auto-fluorescence of the biological sample and the sample substrate [3]. Suppressing this fluorescence background would enhance the contrast between the diseased cells and normal cells, allowing better classification, which is desirable for clinical applications [3,4,5]. Numerous techniques have been demonstrated to reduce or suppress fluorescence background [6,7,8,9,10,11]. The general principle of this technique was demonstrated three decades ago [12,13,14,15], it was only recently that this method has evolved and been successfully adapted for biomedical applications [3]

Methods

Results

Conclusion