Removal of cross-phase modulation artifacts in ultrafast pump–probe dynamics by deep learning

Abstract

Removing non-resonant background from CARS spectra via deep learning APL Photonics 5, 061305 (2020);

Computational coherent Raman scattering imaging: breaking physical barriers by fusion of advanced instrumentation and data science

“…Deep-learning-based background removal is advantageous in that it requires no experimental measurement of the non-resonant background and has a fast processing speed of milliseconds. A similar approach has also been demonstrated to remove cross-phase modulation background in spectroscopic SRS [114] using the peak width differences between the Raman band and the background.…”

“…Compressive sensing Berto [68] Takizawa [69] Matrix completion Lin [59] Supervised spectral sub-sampling Freudiger [71], Bae [72], Pence [74] Masia [73] Digital holography Shi [78], Cocking [79] Projection tomography Chen [80], Lin [81], Gong [85] Deep learning denoising Manifold [97], Lin [36], Abdolghader [100] Yamato [98], Vernuccio [101] Deep learning segmentation & Clinical decision making Orringer [104], Hollon [105], Zhang [106], Feizpour [107] Manuscu [108], Aljakouch [109], Weng [110] Deep learning background removal Bresci [114] Houhou [111], Valensise [112], Wang [113] Deep learning chemical maps prediction Zhang [56], Liu [118], Manifold [115] as existing methods remain viable to boost the newly established design space, and new methods may arise to achieve breakthroughs in aspects such as field of view, imaging depth, and spatial resolution.…”

Artificial Intelligence in Classical and Quantum Photonics

“…In this framework, Bresci et al. [ 68 ] reported an AI‐driven model to retrieve pump–probe ultrafast electronic dynamics embedded in XPM artifacts. The CNN model, “XPMnet,” was trained on $$10^5$$ inputs made up of data‐augmented experimentally measured XPM artifacts superimposed on simulated exponential electronic cooling dynamics.…”

“…That is why the development of efficient methods to tackle the issue of XPM artifact removal urges, but the literature on the topic is somewhat restricted. In this framework, Bresci et al [68] reported an AIdriven model to retrieve pump-probe ultrafast electronic dynamics embedded in XPM artifacts. The CNN model, "XPMnet," was trained on 10 5 inputs made up of data-augmented experimentally measured XPM artifacts superimposed on simulated exponential electronic cooling dynamics.…”

Ultrafast broadband optical modulation in indium tin oxide/titanium dioxide 1D photonic crystal

“…Around zero time delay we can observe the so-called cross phase modulation artifact (XPM), a coherent artifact that originates from the redistribution of the spectral component of the probe induced by the Kerr effect, a change in the refractive index caused by the strong pump pulse [31,32]. Such fast change in the refractive index occurs only when pump and probe overlap in time on the glass substrate and is responsible for the initial positive-negativepositive signal [33]. As we discussed above the relaxation dynamics are faster than 1 ps and can be seen in Figure 2a.…”