Microsecond fingerprint stimulated Raman spectroscopic imaging by ultrafast tuning and spatial-spectral learning

Abstract

Label-free vibrational imaging by stimulated Raman scattering (SRS) provides unprecedented insight into real-time chemical distributions. Specifically, SRS in the fingerprint region (400–1800 cm−1) can resolve multiple chemicals in a complex bio-environment. However, due to the intrinsic weak Raman cross-sections and the lack of ultrafast spectral acquisition schemes with high spectral fidelity, SRS in the fingerprint region is not viable for studying living cells or large-scale tissue samples. Here, we report a fingerprint spectroscopic SRS platform that acquires a distortion-free SRS spectrum at 10 cm−1 spectral resolution within 20 µs using a polygon scanner. Meanwhile, we significantly improve the signal-to-noise ratio by employing a spatial-spectral residual learning network, reaching a level comparable to that with 100 times integration. Collectively, our system enables high-speed vibrational spectroscopic imaging of multiple biomolecules in samples ranging from a single live microbe to a tissue slice.

Deep learning autofluorescence-harmonic microscopy

“…Due to the increasing computing power and quantity of available data, a variety of deep learning methods, ranging from early convolutional neural networks (CNNs) to recent promising generative adversarial networks (GANs), have been proposed and have shown great accomplishments in biomedical imaging 13 – 20 . This significant progress includes super-resolution 13 , 15 , 16 , medical diagnosis 21 , cellular component classification 17 , and virtual H&E staining 14 , 22 .…”

LIPA-driven hydrolysis of cholesteryl arachidonate promotes cancer metastasis via NF-κB

“…Consistent with the fs-SRS imaging results, the abundant LDs in Lalistat 1-treated T24 cells were verified by hSRS. On basis of spectral references, the C-H hyperspectral stacks were decomposed into chemical maps of protein, TAG, and CE by applying pixel-wise LASSO unmixing analysis 26 . For the protein channel, the signals are distributed all over the cell with an enhanced contrast for protein-rich organelles, such as the endoplasmic reticulum and nucleolus.…”

“…No cell damage was observed during the imaging procedure. Least absolute shrinkage and selection operator (LASSO) unmixing process was done following previously published protocol 26 . This method separates different chemicals in the same window based on their spectral profiles.…”

Bacterial growth monitored by two-dimensional tandem mass spectrometry

“…But having chosen lipids and the scale of the experiment, as well as the need to measure many related molecules, only methods based on mass spectrometry (MS) need be considered further. Spectroscopic measurements can broadly detect lipid composition, and in some cases follow individual molecular species in 3D 8 and in vivo 9–11 and they have naturally excellent time resolution, 12,13 but they fall short in their broad applicability in specific molecular recognition. The most widely used MS method of assessing the lipid content of a cell is by liquid chromatography tandem mass spectrometry (LC-MS/MS).…”