Published in Biomedical Optics Express, 2025
Introduce a novel approach that combines ptychographic scanning along a spiral path with the ePIE algorithm, enabling accurate reconstruction of the original image.
Recommended citation: Chen, X., H. Sha, C. Chen, Y. Jiang, W. Zou and Y. Zhang (2025). "Adaptive lensless microscopic imaging with unknown phase modulation." Biomedical Optics Express 16(3).
Download Paper
Published in Biosensors and Bioelectronics, 2025
Introduce a novel approach that combines ptychographic scanning along a spiral path with the ePIE algorithm, enabling accurate reconstruction of the original image.
Recommended citation: Zheng H, et al. Rational development of Nile red derivatives with significantly improved specificity and photostability for advanced fluorescence imaging of lipid droplets. Biosensors and Bioelectronics 282, (2025).
Download Paper
Published in Nature Computational Science, 2024
We present a general framework for evaluating deep-learning-based OOD dynamics-detection methods. We further develop a baseline approach that achieves robust OOD dynamics detection as well as accurate recognition of in-distribution anomalous diffusion. We demonstrate that this method enables a reliable characterization of complex behaviors across a wide range of experimentally diverse systems.
Recommended citation: Feng, X., Hao, Sha., et al. Reliable deep learning in anomalous diffusion against out-of-distribution dynamics. Nature Computational Science 4(2024).
Download Paper
Published in Communications Engineering, 2024
We build a novel deep compressive confocal microscope (DCCM). Combined with a deep learning reconstruction algorithm (DCCM-Net), our DCCM is able to achieve high-quality confocal microscopic imaging with low phototoxicity. Our imaging experiments with fluorescent microspheres demonstrate its capability of achieving single-pixel confocal imaging with a sampling ratio of only approximately 0.05%. In the prototype of our DCCM system, the imaging rates can achieve up to 833 fps and 13.65 megapixel/s.
Recommended citation: Liu, S., et al. Compressive confocal microscopy imaging at the single-photon level with ultra-low sampling ratios. Communications Engineering 3(2024).
Download Paper
Published in Optics Express, 2024
We propose a snapshot hyperspectral confocal microscopic imaging system (SHCMS). It combined coded illumination microscopy based on a digital micromirror device (DMD) with a snapshot hyperspectral confocal neural network (SHCNet) to realize single-shot confocal hyperspectral imaging. With SHCMS, high-contrast 160-band confocal hyperspectral images of potato tuber autofluorescence can be collected by only single-shot, which is almost 5 times improvement in the number of spectral channels than previously reported methods.
Recommended citation: Liu, S., et al. Deep learning-enhanced snapshot hyperspectral confocal microscopy imaging system. Optics Express 32(2024).
Download Paper
Published in APL Photonics, 2023
A deep learning-enhanced single-molecule spectrum imaging method (SpecGAN) for improving the single-molecule spectrum imaging efficiency. With SpecGAN, the super-resolution spectrum image of the COS-7 membrane can be reconstructed with merely 12,000 frames of single-molecule localization images, which is almost half of the previously reported frame count for spectrally resolved super-resolution imaging.
Recommended citation: Sha, H., Li, H., Zhang, Y. & Hou, S. Deep learning-enhanced single-molecule spectrum imaging. APL Photonics 8(2023).
Download Paper
Published in Biomedical Optics Express, 2023
Propose a deep learning-based blindly spectral unmixing method, termed AutoUnmix, to imitate the physical spectral mixing process. Our proposed method has demonstrated real-time unmixing capabilities, surpassing existing methods by up to 100-fold in terms of unmixing speed.
Recommended citation: Jiang, Y., Sha, H., Liu, S., Qin, P. & Zhang, Y. AutoUnmix: an autoencoder-based spectral unmixing method for multi-color fluorescence microscopy imaging. Biomedical Optics Express 14(2023).
Download Paper
Published in Optics and Lasers in Engineering, 2022
A deep learning-enhanced single-molecule spectrum imaging method (SpecGAN) for improving the single-molecule spectrum imaging efficiency. With SpecGAN, the super-resolution spectrum image of the COS-7 membrane can be reconstructed with merely 12,000 frames of single-molecule localization images, which is almost half of the previously reported frame count for spectrally resolved super-resolution imaging.
Recommended citation: Liu, S., et al. Intensity and phase imaging through scattering media via deep despeckle complex neural networks. Optics and Lasers in Engineering 159(2022).
Download Paper
Published in Laser & Optoelectronics Progress, 2021
An algorithm combining computational imaging prior knowledge and deep learning is proposed to deal with the FPM reconstruction process.
Recommended citation: Sha hao, Liu Yangzhe & Zhang Yongbing. 基于深度学习的傅里叶叠层成像技术. Laser & Optoelectronics Progress 58(2021).
Download Paper