2025/02/11 by Ashesh Ashesh, Federico Carrara, Igor Zubarev +23 路 1 voice 路 5 citations
Biochemistry, Genetics and Molecular Biology 路 #Artificial intelligence #Cell Image Analysis Techniques #Computer science #Fluorescence #Fluorescence microscope #Microscopy #Natural language processing #Optics #Physics
paper 路 doi:10.1101/2025.02.10.637323
published in bioRxiv (Cold Spring Harbor Laboratory) (Cold Spring Harbor Laboratory)
openalex created_date 2025/02/11 路 openalex publication_date 2025/02/11 路 openalex updated_date 2026/07/30
Abstract Fluorescence microscopy, a key driver for progress in the life sciences, faces limitations due to the microscope鈥檚 optics, fluorophore chemistry, and photon exposure limits, necessitating trade-offs in imaging speed, resolution, and depth. Here, we introduce Micro饾晩plit, a computational multiplexing technique based on deep learning that allows multiple cellular structures to be imaged in a single fluorescent channel and then unmix them by computational means, allowing faster imaging and reduced photon exposure. We show that Micro饾晩plit efficiently separates up to four superimposed noisy structures into distinct denoised fluorescent image channels. Furthermore, using Variational Splitting Encoder-Decoder (VSE) networks, our approach can sample diverse predictions from a trained posterior of solutions. The diversity of these samples scales with the uncertainty in a given input, allowing us to estimate the true prediction errors by computing the variability between posterior samples. We demonstrate the robustness of Micro饾晩plit networks, which are trained for each splitting task at hand, across various datasets and noise levels and show its utility to image more, to image faster, and to improve downstream analysis. We provide Micro饾晩plit along with all associated training and evaluation datasets as open resources, enabling life scientists to immediately benefit from the potential of computational multiplexing and thus help accelerate the rate of their scientific discovery process.