Bin Duan

The paper introduces a parallelly distributed microscope and software system for high-throughput multispectral 3D imaging with high resolution and data throughput.

The paper develops a platform for storing, visualizing, and analyzing large-scale 3D images in HPC environments using a scalable storage format and web-based annotation tools.

This paper develops a compression method that achieves high compression ratios with minimal visual and quantitative errors for single-molecule imaging data.

This paper introduces a method using non-local correlation in the embedding space to improve cell instance segmentation in EM images, reducing over- and under-segmentation issues.

This paper introduces an unsupervised method for tracing neurons in multispectral Brainbow images, achieving more accurate results than existing methods by reducing segmentation and continuity errors.