Diffusion-Based Inference for Heterogeneous Multichannel Fluorescence Imaging
- 1Institut de Recerca Sant Joan de Déu
Abstract
Fluorescence microscopy is a central tool in biomedical image analysis, but its practical use is often constrained by incomplete and heterogeneous data. This is especially true in rare disease research, where patient cohorts are small, biological material is limited, and complex study designs frequently yield images with missing, degraded, or low-quality channels. We introduce Reliability-Mask-Conditioned DDPM (RM-DDPM), a conditional diffusion model that formulates multichannel image generation, missing-channel imputation, and degraded-channel restoration as a single reliability-guided inference problem. The model is trained on a curated subset of complete, high-quality multichannel images using a self-supervised masking objective to learn rich cross-channel dependencies from the available channels. At inference time, reliable observations condition the reverse diffusion process, allowing the same trained model to solve multiple reconstruction tasks without task-specific retraining. We evaluate RM-DDPM against task-specific baselines on one public and two in-house fluorescence microscopy datasets, comprising a rare-disease preclinical model study and a clinical patient cohort. Across these settings, RM-DDPM matches the specialized baselines in reconstruction quality, hallucinates fewer false structures, and remains more robust as degradation severity increases, while repeated sampling provides per-pixel uncertainty maps that flag unreliable reconstructions.
Related
- Fluorescence Correlation Spectroscopy To Study Diffusion and Reaction of Bacteriophages inside Biofilms
- Quanta microscopy enables fluorescence imaging at the native bit-plane rate
- Multi-parametric NIR-II fluorescence perfusion imaging towards quantitative stroke evaluation
- Optimizing x-ray based multimodal imaging: investigating dose dependency and modulation of fluorescence emission from GdF ₃ nanoparticles doped with terbium using spectral photon counting CT
- Live-Imaging of Molecular Processes with Real-Time Fluorescence Microscopy
BibTeX
@misc{georvasilis2026diffusionbased,
title = {{Diffusion-Based Inference for Heterogeneous Multichannel Fluorescence Imaging}},
author = {I. Georvasilis and M. Frias Nestares and C. Badosa and U. Musokhranova and A. Oyarzabal and C. Jimenez-Mallebrera and J. Ruminski and M. Roldan},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.07.753438},
url = {https://www.biorxiv.org/content/10.64898/2026.10.07.753438v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Diffusion-Based Inference for Heterogeneous Multichannel Fluorescence Imaging
AU - I. Georvasilis
AU - M. Frias Nestares
AU - C. Badosa
AU - U. Musokhranova
AU - A. Oyarzabal
AU - C. Jimenez-Mallebrera
AU - J. Ruminski
AU - M. Roldan
PY - 2026
DA - 2026/10/07
PB - bioRxiv
DO - 10.64898/2026.10.07.753438
UR - https://www.biorxiv.org/content/10.64898/2026.10.07.753438v1
AB - Fluorescence microscopy is a central tool in biomedical image analysis, but its practical use is often constrained by incomplete and heterogeneous data. This is especially true in rare disease research, where patient cohorts are small, biological material is limited, and complex study designs frequently yield images with missing, degraded, or low-quality channels. We introduce Reliability-Mask-Conditioned DDPM (RM-DDPM), a conditional diffusion model that formulates multichannel image generation, missing-channel imputation, and degraded-channel restoration as a single reliability-guided inference problem. The model is trained on a curated subset of complete, high-quality multichannel images using a self-supervised masking objective to learn rich cross-channel dependencies from the available channels. At inference time, reliable observations condition the reverse diffusion process, allowing the same trained model to solve multiple reconstruction tasks without task-specific retraining. We evaluate RM-DDPM against task-specific baselines on one public and two in-house fluorescence microscopy datasets, comprising a rare-disease preclinical model study and a clinical patient cohort. Across these settings, RM-DDPM matches the specialized baselines in reconstruction quality, hallucinates fewer false structures, and remains more robust as degradation severity increases, while repeated sampling provides per-pixel uncertainty maps that flag unreliable reconstructions.
ER -