DeepSTORM3D: dense 3D localization microscopy and PSF design by deep learning.

A team of researchers from the Technion have developed an innovative microscopic method to create dynamic 3D images based on deep learning. The team included Elias Nehme, a PhD student in Electrical Engineering Faculty, and his supervisors, Dr. Yoav Shechtman from the Faculty of Biomedical Engineering and Dr. Tomer Michaeli from the Faculty of Electrical Engineering.

Their approach, DeepSTORM3D, enables the study of biological processes in whole cells at timescales that are rarely explored in localization microscopy.

The researchers demonstrated their approach experimentally with super-resolution reconstructions of mitochondria and volumetric imaging of fluorescently labeled telomeres in cells.

The research was conducted in collaboration with Google and the RBNI center in the Technion.,7340,L-5749252,00.html