平场复用多焦点结构光照明超分辨显微成像  

Flat-field multiplexed multifocal structured illumination super-resolution microscopy

作　　者：葛阳阳 何灼奋 黄黎琳 林丹樱[1] 曹慧群 屈军乐[1] 于斌[1] Ge Yang-Yang;He Zhuo-Fen;Huang Li-Lin;Lin Dan-Ying;Cao Hui-Qun;Qu Jun-Le;Yu Bin(Key Laboratory of Optoelectronic Devices and Systems,Ministry of Education and Guangdong Province,College of Physics and Optoelectronic Engineering,Shenzhen University,Shenzhen 518060,China;College of Chemistry and Environmental Engineering,Shenzhen University,Shenzhen 518060,China)

机构地区：[1]深圳大学物理与光电工程学院,光电子器件与系统教育部/广东省重点实验室,深圳518060 [2]深圳大学化学与环境工程学院,深圳518060

出　　处：《物理学报》2022年第4期343-351,共9页Acta Physica Sinica

基　　金：国家自然科学基金(批准号:61975131,62175166,61835009,61775144,);深圳市基础研究项目(批准号:JCYJ20200109105411133,JCYJ20170412105003520)资助的课题.

摘　　要：多焦点结构光照明显微技术(multifocal structured illumination microscopy,MSIM)能在50μm的成像深度内和1Hz的成像速度下实现两倍于衍射极限分辨率的提升,相比传统的宽场结构光照明显微技术,具有较大的成像深度和层析能力,更适合应用于厚样品的长时程三维超分辨成像.然而,MSIM存在成像速度慢、图像处理过程复杂等问题.本文提出了一种基于平场复用多焦点结构光照明的快速超分辨显微成像方法和系统(flat-field multiplexed MSIM,FM-MSIM),通过在照明光路中插入光束整形器件,将高斯光束转变为均为分布的平顶光束,提高激发点阵的强度均匀性和扩大视场;通过将每个衍射受限的激发点沿y方向延长,形成新的多路复用多焦点阵照明图案,提高能量利用率,减少扫描步数,进而提高成像速度和信噪比;结合基于多重测量矢量模型的稀疏贝叶斯学习图像重构算法,简化图像重构步骤,在保证空间分辨率的同时实现至少4倍于传统MSIM的成像速度.在此基础上,利用搭建的FM-MSIM系统进行了BSC细胞微管样片和小鼠肾切片标准样片的超分辨成像实验,实验结果证明了该系统的快速三维超分辨成像能力,对于MSIM的发展具有重要的意义.Multifocal structured illumination microscopy(MSIM)can achieve optically sectioned images with twice the diffraction limited resolution at an imaging speed of 1 Hz and an imaging depth of up to 50μm.Compared with the traditional wide-field SIM,the MSIM has greater imaging depth and optical sectionning ability,and it is more suitable for long-term three-dimensional(3D)super-resolution imaging of living thick samples.However,the MSIM has some problems,such as slow imaging speed and complex image post-processing process.In this work,a fast super-resolution imaging method and system based on the flat-field multiplexed MSIM(FM-MSIM)is proposed.By inserting a beam shaping device into the illumination light path,the Gaussian beam is reshaped into a uniform flat-top profile,thereby improving the intensity uniformity of excitation multi-spot focal array and expanding the field of view.By elongating each diffraction limited excitation focal point four times along the Y direction to form a new multiplexed multifocal array pattern,the number of scanning steps is reduced,the energy utilization is improved,and then the imaging speed and signal-to-noise ratio are improved.Combined with the sparse Bayesian learning image reconstruction algorithm based on multiple measurement vector model,the image reconstruction steps are simplified,the imaging speed can be improved at least 4 times while ensuring the spatial resolution of MSIM.On this basis,the established FM-MSIM system is used to carry out the super-resolution imaging experiments on the BSC cell microtubule samples and mouse kidney slices.The experimental results prove the fast three-dimensional super-resolution imaging ability of the system,which is of great significance in developing the fast MSIM.

关 键 词：多焦点结构光照明显微技术 超分辨成像 平场照明 贝叶斯学习算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]