Optimal transport unlocks end-to-end learning for single-molecule localization

ICLR 2026 Conference Submission16705 Authors

Keywords: Single Molecule Localization Microscopy, SMLM, High-Density, Learning, Deep Learning, Inverse Problems, Iterative Refinement

TL;DR: We present an end-to-end learning model for single-molecule localization microscopy that combines an optimal transport loss function with an iterative neural network. It achieves state-of-the-art performance on synthetic benchmarks and real images.

Abstract:

Single‑molecule localization microscopy (SMLM) surpasses the diffraction limit by detecting and localizing individual fluorophores - fluorescent molecules stained onto the observed specimen - over time to reconstruct super‑resolved images. Conventional SMLM requires non‑overlapping emitting fluorophores, leading to long acquisition times that hinders live‑cell imaging. Although recent deep‑learning approaches can handle denser emissions, they rely on non‑maximum suppression (NMS) layers, which are non‑differentiable and may discard true positives with their local fusion strategy.
In this presentation, we reformulate the SMLM training objective as a set‑matching problem, deriving an optimal‑transport loss that eliminates the need for NMS during inference and enables end‑to‑end training.
Additionally, we propose an iterative neural network that integrates knowledge of the microscope’s optical system inside our model.
Experiments on synthetic benchmarks and real biological data show that both our new loss function and architecture surpass the state of the art at moderate and high emitter densities. Code and data are provided in the supplementary material.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Submission Number: 16705