<math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mi>Micro</mi> <mi>S</mi> <mi>plit</mi></mrow> </math> : semantic unmixing of fluorescent microscopy data.

Abstract

Fluorescence microscopy is constrained by optical limits, fluorophore chemistry and finite photon budgets, imposing trade-offs between imaging speed, resolution and phototoxicity. Here we introduce <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mi>Micro</mi> <mi>S</mi> <mi>plit</mi></mrow> </math> , a deep learning-based computational multiplexing method that enables multiple cellular structures to be imaged simultaneously in a single fluorescent channel and then computationally unmixed. We show that <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mspace></mspace> <mrow><mrow><mi>M</mi> <mi>i</mi> <mi>c</mi> <mi>r</mi> <mi>o</mi></mrow> </mrow> <mrow><mrow><mi>S</mi></mrow> </mrow> <mrow><mrow><mi>p</mi> <mi>l</mi> <mi>i</mi> <mi>t</mi></mrow> </mrow> </mrow> </math> separates up to four superimposed noisy structures into distinct, denoised image channels, enabling faster and more photon-efficient imaging. Built on Variational Splitting Encoder-Decoder networks, <math xmlns="http://www.w3.org/1998/Math/MathML"> <mrow><mrow><mi>M</mi> <mi>i</mi> <mi>c</mi> <mi>r</mi> <mi>o</mi></mrow> </mrow> <mrow><mrow><mi>S</mi></mrow> </mrow> <mrow><mrow><mi>p</mi> <mi>l</mi> <mi>i</mi> <mi>t</mi></mrow> </mrow> </math> models a posterior distribution over solutions, allowing uncertainty-aware predictions and the estimation of spatially resolved prediction errors from posterior variability. We demonstrate robust performance across diverse datasets, noise levels and imaging conditions, and show that <math xmlns="http://www.w3.org/1998/Math/MathML"> <mrow><mrow><mi>M</mi> <mi>i</mi> <mi>c</mi> <mi>r</mi> <mi>o</mi></mrow> </mrow> <mrow><mrow><mi>S</mi></mrow> </mrow> <mrow><mrow><mi>p</mi> <mi>l</mi> <mi>i</mi> <mi>t</mi></mrow> </mrow> </math> improves downstream analysis while reducing photon exposure. All methods, data and trained models are released as open resources, enabling immediate adoption of computational multiplexing in biological imaging.