Descattering and image restoration with a transformer-based neural network in deep tissue imaging

• 成果类型: Article
• 署名作者: Xu, Xiangcong; Zhang, Renlong; Luo, Chenggui; Zhang, Chi; Li, Yanping; Lin, Danying; Yu, Bin; Liu, Liwei; Weng, Xiaoyu; Wang, Yiping; Kong, Lingjie; Li, Jia; Qu, Junle
• 署名单位: Shenzhen University; Tsinghua University
• 刊物名称: PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
• ISSN/ISSBN: 0027-8424; 1091-6490
• DOI: 10.1073/pnas.2503576122
• 发表日期: 2025-10-28
• 页码: e2503576122
• 关键词: de-scattering; Deep learning; two-photon excitation fluorescence microscopy; attention mechanism; deep tissue imaging

摘要

Imaging biological structures deep inside tissues is crucial but challenging due to common light scattering. This study proposes a multiattention network that directly maps degraded scattering two-photon excitation fluorescence (TPEF) images to high-quality scattering-free images, thereby computationally extending the imaging depth for TPEF without requiring complex optical additions. The model relies solely on simulated data rather than well-registered real data pairs, and is trained to descatter and restore hidden spatial information at greater depths. Quantitative evaluations on simulated fluorescent beads and vasculature show significant performance improvements in peak signal-to-noise ratio (23 to 29 dB) and structural similarity index (23x) compared to the raw data. We also apply the framework to various ex vivo and in vivo experiments, achieving clear visualization of lipid droplets up to a depth of 1,300 mu m and of vascular structure and astrocytes up to 950 mu m and 500 mu m, respectively, in live mouse brains at lower excitation powers.