Vision Transformers as a Framework for Imaging Fluorescence Correlation Spectroscopy
- 1National University of Singapore
Abstract
Fluorescence correlation spectroscopy (FCS) is widely used for quantifying molecular dynamics in live cells. However, conventional FCS relies on the autocorrelation function (ACF), a lossy transformation that discards higher-order information, mathematically couples distinct biophysical parameters, and requires long measurements (~50,000 frames, or ~1 minute) that obscure short-lived transient events. To overcome this information bottleneck, we introduce ImFCSViT, a spatiotemporal Vision Transformer-based framework that bypasses the ACF representation entirely, operating directly on raw fluorescence image stacks. By introducing a dedicated normalization token to preserve global intensity statistics, our approach simultaneously extracts four biophysical parameters (diffusion coefficient D, particle density N, particle brightness B, and PSF structure factor {sigma}0) in a single forward pass. Requiring only 2.5 s of data, and potentially down to 1 s (a 1-2 order reduction in data requirement), our approach accurately recovers diffusion coefficients while utilizing its spatial field-of-view to reveal spatially heterogeneous, slow-diffusing sub-populations in lipid bilayers and live-cell plasma membranes that are invisible to classical FCS. By leveraging the data-driven representational learning capabilities of deep learning, we present a flexible, generalizable, and powerful tool for mapping complex, multi-parameter spatiotemporal dynamics in biological systems.
Related
- siRNA/mRNA payload of lipid nanoparticles quantified using fluorescence microscopy and correlation spectroscopy
- Distinguishing Free Cytosolic and Vesicle-Confined Nanoparticle Motion by Fluorescence Correlation Spectroscopy.
- Monitoring intramolecular dynamics across two regions of the mouse prion protein during misfolding and oligomerization using fluorescence correlation spectroscopy
- A computational super-resolution framework for multidimensional fluorescence imaging
- Fluorescence imaging-induced phospholipid oxidation drives membrane phase separation by shifting miscibility boundaries
BibTeX
@misc{sim2026vision,
title = {{Vision Transformers as a Framework for Imaging Fluorescence Correlation Spectroscopy}},
author = {S. R. Sim and Z. Wang and S. Zhou and S. Pandey and T. Wohland},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.21.760315},
url = {https://www.biorxiv.org/content/10.64898/2026.10.21.760315v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Vision Transformers as a Framework for Imaging Fluorescence Correlation Spectroscopy
AU - S. R. Sim
AU - Z. Wang
AU - S. Zhou
AU - S. Pandey
AU - T. Wohland
PY - 2026
DA - 2026/10/21
PB - bioRxiv
DO - 10.64898/2026.10.21.760315
UR - https://www.biorxiv.org/content/10.64898/2026.10.21.760315v1
AB - Fluorescence correlation spectroscopy (FCS) is widely used for quantifying molecular dynamics in live cells. However, conventional FCS relies on the autocorrelation function (ACF), a lossy transformation that discards higher-order information, mathematically couples distinct biophysical parameters, and requires long measurements (~50,000 frames, or ~1 minute) that obscure short-lived transient events. To overcome this information bottleneck, we introduce ImFCSViT, a spatiotemporal Vision Transformer-based framework that bypasses the ACF representation entirely, operating directly on raw fluorescence image stacks. By introducing a dedicated normalization token to preserve global intensity statistics, our approach simultaneously extracts four biophysical parameters (diffusion coefficient D, particle density N, particle brightness B, and PSF structure factor {sigma}0) in a single forward pass. Requiring only 2.5 s of data, and potentially down to 1 s (a 1-2 order reduction in data requirement), our approach accurately recovers diffusion coefficients while utilizing its spatial field-of-view to reveal spatially heterogeneous, slow-diffusing sub-populations in lipid bilayers and live-cell plasma membranes that are invisible to classical FCS. By leveraging the data-driven representational learning capabilities of deep learning, we present a flexible, generalizable, and powerful tool for mapping complex, multi-parameter spatiotemporal dynamics in biological systems.
ER -