Fast denoising and lossless spectrum extraction in stimulated Raman scattering microscopy

Abstract

Stimulated Raman scattering (SRS) microscopy is a nonlinear optical imaging method for visualizing chemical content based on molecular vibrational bonds. However, the imaging speed and sensitivity are currently limited by the noise of the light beam probing the Raman process. In this paper, we present a fast non‐average denoising and high‐precision Raman shift extraction method, based on a self‐reinforcing signal‐to‐noise ratio (SNR) enhancement algorithm, for SRS spectroscopy and microscopy. We compare the results of this method with the filtering methods and the reported experimental methods to demonstrate its high efficiency and high precision in spectral denoising, Raman peak extraction and image quality improvement. We demonstrate a maximum SNR enhancement of 10.3 dB in fixed tissue imaging and 11.9 dB in vivo imaging. This method reduces the cost and complexity of the SRS system and allows for high‐quality SRS imaging without use of special laser, complicated system design and Raman tags.

Pinpointing Fat Molecules: Advances in Coherent Raman Scattering Microscopy for Lipid Metabolism

“…To address this challenge, several approaches have been developed. Denoising algorithms , have been widely used to improve the signal-to-noise ratio (SNR) of SRS images. For example, Lin et al developed a spatial-spectral residual learning network to enhance the SNR of fingerprint region SRS images.…”

Noises investigations and image denoising in femtosecond stimulated Raman scattering microscopy

“…In Reference 33, to solve this issue in SRS imaging, a denoising algorithm, removing the noise in both spatial and spectral domains by total variation minimization, was demonstrated. In Reference 34, a fast nonaverage denoising and high‐precision Raman shift extraction method, based on a self‐reinforcing SNR enhancement algorithm, was reported for SRS spectroscopy and microscopy. Recently, denoising of SRS microscopy images via deep learning was investigated [35–37].…”

“…[29], in which the optimization of the ultimate SRS detection limit in a conventional lock-inbased SRS system was investigated. Concerning image processing, it is mainly devoted to denoising signals [33][34][35][36][37]. However, the link between hardware characterization and imaging processing often is not highlighted.…”