Search

Noise manifests ubiquitously in nonlinear spectroscopy, where multiple sources contribute to experimental signals generating interrelated unwanted components, from random point-wise fluctuations to structured baseline signals. Mitigating strategies are usually heuristic, depending on subjective biases such as the setting of parameters in data analysis algorithms and the removal order of the unwanted components. We propose a data-driven frequency-domain denoiser based on a convolutional neural network to extract authentic vibrational features from a nonlinear background in noisy spectroscopic raw data. The different spectral scales in the problem are treated in parallel by means of filters with multiple kernel sizes, which allow the receptive field of the network to adapt to the informative features in the spectra. We test our approach by retrieving asymmetric peaks in stimulated Raman spectroscopy, an ideal test-bed due to its intrinsic complex spectral features combined with a strong background signal. By using a theoretical perturbative toolbox, we efficiently train the network with simulated datasets resembling the statistical properties and lineshapes of the experimental spectra. The developed algorithm is successfully applied to experimental data to obtain noise- and background-free stimulated Raman spectra of organic molecules and prototypical heme proteins.

Spontaneous Raman microscopy reveals the chemical composition of a sample in a label-free and non-invasive fashion by directly measuring the vibrational spectra of molecules. However, its extremely low cross section prevents its application to fast imaging. Stimulated Raman scattering (SRS) amplifies the signal by several orders of magnitude thanks to the coherent nature of the nonlinear process, thus unlocking high-speed microscopy applications that provide analytical information to elucidate biochemical mechanisms with subcellular resolution. Nevertheless, in its standard implementation, narrowband SRS provides images at only one frequency at a time, which is not sufficient to distinguish constituents with overlapping Raman bands. Here, we report a broadband SRS microscope equipped with a home-built multichannel lock-in amplifier simultaneously measuring the SRS signal at 32 frequencies with integration time down to 44 µs, allowing for detailed, high spatial resolution mapping of spectrally congested samples. We demonstrate the capability of our microscope to differentiate the chemical constituents of heterogeneous samples by measuring the relative concentrations of different fatty acids in cultured hepatocytes at the single lipid droplet level and by differentiating tumor from peritumoral tissue in a preclinical mouse model of fibrosarcoma.