Deep learning speeds up electron microscopy brain imaging dramatically
ZoomSEM: accelerating large-scale brain connectomics with intelligent SEM super-resolution
Computer Vision and Pattern Recognition
Summary
Imaging the brain in extremely high detail using electron microscopes takes a very long time, sometimes months or years. The authors created ZoomSEM, a smart computer method that takes quicker, lower-resolution scans and turns them into much sharper images, preserving tiny brain details needed to study how neurons connect. This method can make the scanning process up to 30 times faster. It works well on various biological samples, promising to speed up brain research and other areas that need detailed imaging.
What this means in practice
- •For brain imaging teams: Produce high-resolution 3D brain images much faster by converting quick low-res scans into detailed images suitable for connectomic analysis.
- •For biomedical imaging developers: Enhance SEM hardware with AI-based super-resolution to accelerate image acquisition while maintaining structural detail for diverse biological samples.
Authors
Jiateng Shou, Zhiwei Xiong, Ruikang Xu, Zhicheng Wang, Peiyao Shi, Te Shi, Haiqun Jin, Chunying Yin, Jun Guo, Manfred Auer, Hanchuan Peng, Ruobing Zhang, Feng Wu
Abstract
Serial electron microscopy (EM) is essential for large-volume biological imaging and connectomics, yet its throughput remains a major bottleneck, with cubic-millimeter-scale brain imaging requiring months or years. Existing deep learning-based EM super-resolution methods are limited to small upscaling factors or rely on synthetic data, offering modest acceleration and limited practical impact. We introduce ZoomSEM, a deep learning-based super-resolution method to dramatically accelerate SEM imaging. Trained on real low- and high-resolution scanning electron microscopy (SEM) pairs, ZoomSEM enables reliable recovery of 4$\times$, 8$\times$, and up to 16$\times$ higher-resolution images from fast low-resolution scans while preserving fine neuronal structures critical for connectomic analysis, achieving up to 30-fold effective acceleration. Extensive experiments across datasets of diverse biological samples demonstrate robust generalization under realistic low-resolution acquisition. This advance enables high-resolution, large-volume serial EM imaging at unprecedented speed, poised to transform connectomics and related fields.