High-Speed Chemical Imaging by Dense-Net Learning of Femtosecond Stimulated Raman Scattering

• DOI: 10.1021/acs.jpclett.0c01598
• Publication Date: 2020-09-11T15:04:02Z
• Authors: Jing Zhang; Jian Zhao; Haonan Lin; Yuying Tan; Ji-Xin Cheng
• Keywords: SIGNAL (programming language); Chirp

Hyperspectral stimulated Raman scattering (SRS) by spectral focusing can generate label-free chemical images through temporal scanning of chirped femtosecond pulses. Yet, pulse chirping decreases the peak power and increases acquisition time, resulting in a much slower imaging speed compared to single-frame SRS using In this paper, we present deep learning algorithm solve inverse problem getting chemically labeled image from image. Our DenseNet-based method, termed as DeepChem, achieves high-speed with large signal level. Speed is improved 2 orders magnitude four subcellular components (lipid droplet, endoplasmic reticulum, nuclei, cytoplasm) classified MIA PaCa-2 cells other cell types which were not used for training. Lipid droplet dynamics cellular response dithiothreitol live are demonstrated computationally multiplex method.