Laser scanning microscopy has inherent tradeoffs between imaging speed, field of view (FOV), and spatial resolution due to the limitations of sophisticated mechanical and optical setups, and deep learning networks have emerged to overcome these limitations without changing the system. Here, we demonstrate deep learning autofluorescence-harmonic microscopy (DLAM) based on self-alignment attention-guided residual-in-residual dense generative adversarial networks to close the gap between speed, FOV, and quality. Using the framework, we demonstrate label-free large-field multimodal imaging of clinicopathological tissues with enhanced spatial resolution and running time advantages. Statistical quality assessments show that the attention-guided residual dense connections minimize the persistent noise, distortions, and scanning fringes that degrade the autofluorescence-harmonic images and avoid reconstruction artifacts in the output images. With the advantages of high contrast, high fidelity, and high speed in image reconstruction, DLAM can act as a powerful tool for the noninvasive evaluation of diseases, neural activity, and embryogenesis.
elastically scattered photons) featuring time-consuming spectral integration and hence a poor acquisition speed (several seconds per line or minutes per frame). This weakness constitutes a large obstacle for real-time monitoring of dynamic events in organisms. Recently, several variations of Raman spectroscopy have been developed to enhance the sensitivity of Raman spectroscopy and dynamic processes of biological samples, including coherent Raman scattering (CRS) spectroscopy and surfaceenhanced Raman spectroscopy (SERS). SERS involves a plasmonic effect where molecules absorbed on a rough metal surface can result in high Raman scattering intensities by increasing the incident electric field. [5] For SERS technology, a SERS substrate must be introduced to enhance the detection sensitivity, and the properties of the substrate (the composition, size, shape and aggregation degree of nanoparticles) affect the shape of the SERS spectra. In addition, only gold, silver, copper, and a few rare alkali metals (such as lithium and sodium) have strong SERS effects, and it is necessary to effectively solve the biocompatibility and safety issues in the preparation of nanomaterials. With the development of laser and nonlinear optics, CRS microscopy has also been demonstrated to break the speed limit for vibrational imaging. [6] CRS has two major forms: coherent anti-Stokes Raman scattering (CARS) and stimulated Raman scattering (SRS); [7] both techniques can dramatically increase the Raman signal by a few orders of magnitude and hence are able to achieve video-rate molecular imaging in vivo. [8,9] CARS and SRS processes often exist in a sample concurrently because of the same excitation conditions (spatiotemporal overlap of the pump and Stokes beam) with a frequency difference (Raman shift) matching the molecular vibrational frequency. However, CARS is an optical parametric process accompanied by a nonresonant part originating from fourwave mixing processes. This intrinsic nonresonant background results in a deviation or even distortion of the Raman spectrum. SRS is a direct process of photon-vibration energy transfer: in brief, the interaction between the electron cloud of a sample and the photons of two lasers creates an induced dipole moment within the molecule based on its polarizability, resulting in a pump energy loss (stimulated Raman loss, SRL) and a Stokes energy increase (stimulated Raman gain, SRG). The Raman signal can, therefore, be deduced from SRL or SRG, resulting in a high-fidelity Raman spectrum free from the nonresonant background. Moreover, the SRS signal exhibits Stimulated Raman scattering (SRS) microscopy is a nonlinear optical imaging method for visualizing chemical content based on molecular vibrational bonds. Featuring high speed, high resolution, high sensitivity, high accuracy, and 3D sectioning, SRS microscopy has made tremendous progress toward biochemical information acquisition, cellular function investigation, and label-free medical diagnosis in the biosciences. In this review, the principle of ...
Many ultrafast phenomena in biology and physics are fundamental to our scientific understanding but have not yet been visualized owing to the extreme speed and sensitivity requirements in imaging modalities. Two examples are the propagation of passive current flows through myelinated axons and electromagnetic pulses through dielectrics, which are both key to information processing in living organisms and electronic devices. Here, we demonstrate differentially enhanced compressed ultrafast photography (Diff-CUP) to directly visualize propagations of passive current flows at approximately 100 m/s along internodes, i.e., continuous myelinated axons between nodes of Ranvier, from Xenopus laevis sciatic nerves and of electromagnetic pulses at approximately 5 × 107 m/s through lithium niobate. The spatiotemporal dynamics of both propagation processes are consistent with the results from computational models, demonstrating that Diff-CUP can span these two extreme timescales while maintaining high phase sensitivity. With its ultrahigh speed (picosecond resolution), high sensitivity, and noninvasiveness, Diff-CUP provides a powerful tool for investigating ultrafast biological and physical phenomena.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.