Statistically unbiased prediction enables accurate denoising of voltage imaging data

Minho Eom, Seungjae Han, Gyuri Kim, Eun‐Seo Cho, Jueun Sim, Pojeong Park, Kang-Han Lee, Seonghoon Kim, Márton Rózsa, Karel Svoboda, Myunghwan Choi, Cheol‐Hee Kim, Adam E. Cohen, Jae‐Byum Chang, Young‐Gyu YoonZenodo 2022 · DOI · OpenAlex

Abstract

Here we report SUPPORT (Statistically Unbiased Prediction utilizing sPatiOtempoRal information in imaging daTa), a self-supervised learning method for removing Poisson-Gaussian noise in voltage imaging data. SUPPORT is based on the insight that a pixel value in voltage imaging data is highly dependent on its spatially neighboring pixels in the same time frame, even when its temporally adjacent frames do not provide useful information for statistical prediction. Such spatiotemporal dependency is captured and utilized to accurately denoise voltage imaging data in which the existence of the action potential in a time frame cannot be inferred by the information in other frames. Through simulation and experiments, we show that SUPPORT enables precise denoising of voltage imaging data while preserving the underlying dynamics in the scene. We also show that SUPPORT can be used for denoising time-lapse fluorescence microscopy images of <em>Caenorhabditis elegans</em> (<em>C. elegans</em>), in which the imaging speed is not faster than the locomotion of the worm, as well as static volumetric images of <em>Penicillium</em> and mouse embryos. SUPPORT is exceptionally compelling for denoising voltage imaging and time-lapse imaging data, and is even effective for denoising calcium imaging data. For more details, please see the accompanying research publication "Statistically unbiased prediction enables accurate denoising of voltage imaging data". Datasets for volumetric structural imaging of penicillium and calcium imaging of zebrafish. <strong>Volumetric structural imaging of <em>penicillium</em></strong><br> Penicillium/penicillium_low_snr.tif<br> --&gt; Low SNR image<br> Penicillium/penicillium_high_snr.tif<br> --&gt; High SNR image <strong>Calcium imaging of zebrafish</strong><br> Zebrafish/zebrafish_multiple_brain_regions.tif<br> --&gt; Multiple brain regions<br> Zebrafish/zebrafish_Cerebellar_plate.tif<br> --&gt; Cerebellar plate<br> Zebrafish/zebrafish_Dorsal_telencephalon.tif<br> --&gt; Dorsal telencephalon<br> Zebrafish/zebrafish_Medulla_oblongata.tif<br> --&gt; Medulla oblongata<br> Zebrafish/zebrafish_Olfactory_bulb.tif<br> --&gt; Olfactory bulb<br> Zebrafish/zebrafish_Optic_tectum.tif<br> --&gt; Optic tectum<br> Zebrafish/zebrafish_Habenula.tif<br> --&gt; Habenula
BibTeX
@inproceedings{eom2022statistically,
  author = {Minho Eom and Seungjae Han and Gyuri Kim and Eun‐Seo Cho and Jueun Sim and Pojeong Park and Kang-Han Lee and Seonghoon Kim and Márton Rózsa and Karel Svoboda and Myunghwan Choi and Cheol‐Hee Kim and Adam E. Cohen and Jae‐Byum Chang and Young‐Gyu Yoon},
  title = {{Statistically unbiased prediction enables accurate denoising of voltage imaging data}},
  booktitle = {Zenodo},
  year = {2022},
  doi = {10.5281/zenodo.7330257},
}

Loading related content...

Figure