Removing non-resonant background from CARS spectra via deep learning

Valensise, Carlo Michele; Giuseppi, Alessandro; Vernuccio, Federico; Cadena, Alejandro De la; Cerullo, Giulio; Polli, Dario

doi:10.1063/5.0007821

Cited by 65 publications

(51 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So-called ghost imaging can be one of such application areas [248]. The DL techniques can also pave the way to reach super-resolution in THz imaging as it evolves in widefield light sheet microscopy, for instance, using an Airy beam [437] or to remove non-resonant background as it was done in CARS experiments [438].…”

Section: Artificial Intelligence In Thz Imagingmentioning

confidence: 99%

Roadmap of Terahertz Imaging 2021

Valušis

Lisauskas

Yuan

et al. 2021

Sensors

178

View full text Add to dashboard Cite

In this roadmap article, we have focused on the most recent advances in terahertz (THz) imaging with particular attention paid to the optimization and miniaturization of the THz imaging systems. Such systems entail enhanced functionality, reduced power consumption, and increased convenience, thus being geared toward the implementation of THz imaging systems in real operational conditions. The article will touch upon the advanced solid-state-based THz imaging systems, including room temperature THz sensors and arrays, as well as their on-chip integration with diffractive THz optical components. We will cover the current-state of compact room temperature THz emission sources, both optolectronic and electrically driven; particular emphasis is attributed to the beam-forming role in THz imaging, THz holography and spatial filtering, THz nano-imaging, and computational imaging. A number of advanced THz techniques, such as light-field THz imaging, homodyne spectroscopy, and phase sensitive spectrometry, THz modulated continuous wave imaging, room temperature THz frequency combs, and passive THz imaging, as well as the use of artificial intelligence in THz data processing and optics development, will be reviewed. This roadmap presents a structured snapshot of current advances in THz imaging as of 2021 and provides an opinion on contemporary scientific and technological challenges in this field, as well as extrapolations of possible further evolution in THz imaging.

show abstract

Section: Artificial Intelligence In Thz Imagingmentioning

confidence: 99%

Roadmap of Terahertz Imaging 2021

Valušis

Lisauskas

Yuan

et al. 2021

Sensors

178

View full text Add to dashboard Cite

show abstract

“…Further development of the CARS technique was aimed at suppressing the nonresonant background affecting CARS measurements. Therefore, to reduce the nonresonant contribution to the signal, researchers have proposed several approaches, including the use of time-resolved CARS [ 55 , 56 ], spectral interferometric polarization CARS [ 57 ], epi-detection of CARS signal [ 58 ], multiplex CARS [ 59 , 60 ], and deep learning models to extract vibrational information from spectra [ 61 ].…”

Section: Introductionmentioning

confidence: 99%

Raman Scattering-Based Biosensing: New Prospects and Opportunities

et al. 2021

View full text Add to dashboard Cite

The growing interest in the development of new platforms for the application of Raman spectroscopy techniques in biosensor technologies is driven by the potential of these techniques in identifying chemical compounds, as well as structural and functional features of biomolecules. The effect of Raman scattering is a result of inelastic light scattering processes, which lead to the emission of scattered light with a different frequency associated with molecular vibrations of the identified molecule. Spontaneous Raman scattering is usually weak, resulting in complexities with the separation of weak inelastically scattered light and intense Rayleigh scattering. These limitations have led to the development of various techniques for enhancing Raman scattering, including resonance Raman spectroscopy (RRS) and nonlinear Raman spectroscopy (coherent anti-Stokes Raman spectroscopy and stimulated Raman spectroscopy). Furthermore, the discovery of the phenomenon of enhanced Raman scattering near metallic nanostructures gave impetus to the development of the surface-enhanced Raman spectroscopy (SERS) as well as its combination with resonance Raman spectroscopy and nonlinear Raman spectroscopic techniques. The combination of nonlinear and resonant optical effects with metal substrates or nanoparticles can be used to increase speed, spatial resolution, and signal amplification in Raman spectroscopy, making these techniques promising for the analysis and characterization of biological samples. This review provides the main provisions of the listed Raman techniques and the advantages and limitations present when applied to life sciences research. The recent advances in SERS and SERS-combined techniques are summarized, such as SERRS, SE-CARS, and SE-SRS for bioimaging and the biosensing of molecules, which form the basis for potential future applications of these techniques in biosensor technology. In addition, an overview is given of the main tools for success in the development of biosensors based on Raman spectroscopy techniques, which can be achieved by choosing one or a combination of the following approaches: (i) fabrication of a reproducible SERS substrate, (ii) synthesis of the SERS nanotag, and (iii) implementation of new platforms for on-site testing.

show abstract

“…However, CARS suffers from a non-resonant background that interferes with the resonant vibrational signal and reduces image contrast, leading to distorted line shapes that decrease the amount of chemical information, resulting in data that is difficult to interpret. 35 In SRS, where the difference in frequency between the ''pump'' and ''Stokes'' photons matches the frequency of a molecular vibration (n vib = n P À n S ), excitation of the vibration is…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%