Abstract:This work develops a deep‐learning‐based algorithm which reconstructs the super‐resolution image directly from raw images of multifocal structured illumination microscopy. The method is validated on microtubules, static and live‐cell mitochondria, and live‐zebrafish at a depth of 100 μm, and achieves state‐of‐the‐art performance with one‐fourths numbers of raw images and one‐third of the runtime required by conventional reconstruction method. Fast and precise reconstruction algorithm is desired for for multifocal structured illumination microscopy (MSIM) to obtain the super‐resolution image. This work proposes a deep convolutional neural network (CNN) to learn a direct mapping from raw MSIM images to super‐resolution image, which takes advantage of the computational advances of deep learning to accelerate the reconstruction. The method is validated on diverse biological structures and in vivo imaging of zebrafish at a depth of 100 μm. The results show that high‐quality, super‐resolution images can be reconstructed in one‐third of the runtime consumed by conventional MSIM method, without compromising spatial resolution. Last but not least, a fourfold reduction in the number of raw images required for reconstruction is achieved by using the same network architecture, yet with different training data.

Deep‐MSIM: Fast Image Reconstruction with Deep Learning in Multifocal Structured Illumination Microscopy