Light-field deep learning enables high-throughput, scattering-mitigated calcium imaging.
Gespeichert in:
| Titel: | Light-field deep learning enables high-throughput, scattering-mitigated calcium imaging. |
|---|---|
| Autoren: | Howe CL; Department of Bioengineering, Imperial College London, Royal School of Mines, London SW7 2AZ, United Kingdom., Zhao KLY; Department of Electrical and Electronic Engineering, Imperial College London, South Kensington, London SW7 2AZ, United Kingdom., Verinaz-Jadan H; Department of Electrical and Electronic Engineering, Imperial College London, South Kensington, London SW7 2AZ, United Kingdom.; Faculty of Electrical and Computer Engineering, Escuela Superior Politécnica del Litoral, Guayaquil 090902, Ecuador., Song P; Department of Electrical and Electronic Engineering, Imperial College London, South Kensington, London SW7 2AZ, United Kingdom.; Department of Engineering, Information Engineering, University of Cambridge, Cambridge CB2 1PZ, United Kingdom., Barnes SJ; Department of Brain Sciences, Division of Neuroscience, United Kingdom Dementia Research Institute, Imperial College London, Hammersmith Hospital Campus, London W12 0NN, United Kingdom., Dragotti PL; Department of Electrical and Electronic Engineering, Imperial College London, South Kensington, London SW7 2AZ, United Kingdom., Foust AJ; Department of Bioengineering, Imperial College London, Royal School of Mines, London SW7 2AZ, United Kingdom. |
| Quelle: | Proceedings of the National Academy of Sciences of the United States of America [Proc Natl Acad Sci U S A] 2025 Dec 02; Vol. 122 (48), pp. e2510337122. Date of Electronic Publication: 2025 Nov 25. |
| Publikationsart: | Journal Article |
| Sprache: | English |
| Info zur Zeitschrift: | Publisher: National Academy of Sciences Country of Publication: United States NLM ID: 7505876 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1091-6490 (Electronic) Linking ISSN: 00278424 NLM ISO Abbreviation: Proc Natl Acad Sci U S A Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: Washington, DC : National Academy of Sciences |
| MeSH-Schlagworte: | Deep Learning* , Calcium*/metabolism, Animals ; Neurons/metabolism ; Mice ; Neocortex/diagnostic imaging ; Neocortex/metabolism ; Image Processing, Computer-Assisted/methods ; Signal-To-Noise Ratio |
| Abstract: | Competing Interests: Competing interests statement:The authors declare no competing interest. Light-field microscopy (LFM) enables high-throughput functional imaging by scanlessly encoding entire volumes in single snapshots. However, LFM's computational burden and vulnerability to scattering limit its application to biological imaging. We present a light-field strategy for volumetric, scattering-mitigated neural circuit activity monitoring. A physics-based deep neural network, 2PiLnet, is trained with two-photon volumes and one-photon light fields. Light-field videos of jGCaMP8f-expressing neurons are acquired in neocortical brain slices. 2PiLnet reconstructs volumes with two-photon-like contrast and source confinement from scattered, blurry one-photon light fields from fields-of-view for which no two-photon images are provided. This enables automated segmentation and extraction of calcium signals with high signal-to-noise ratios and reduces optical crosstalk compared to conventional volume reconstruction methods. Imaging 100 volumes per second, we observe putative spikes fired at up to 10 Hz and the spatial intermingling of putative ensembles throughout [Formula: see text]-micron volumes. Compared to iterative algorithms, 2PiLnet workflows reduce light-field video processing times by several-fold, advancing the goal of real-time, scattering-robust volumetric neural circuit imaging for closed-loop and adaptive experimental paradigms. |
| Grant Information: | 311370/Z/24/Z Wellcome Trust (WT); BB/R009007/1 UKRI | Biotechnology and Biological Sciences Research Council (BBSRC); EP/L016737/1 UKRI | Engineering and Physical Sciences Research Council (EPSRC); EP/W024020/1 UKRI | Engineering and Physical Sciences Research Council (EPSRC); RF1415/14/26 Royal Academy of Engineering (RAENG) |
| Contributed Indexing: | Keywords: calcium imaging; deep learning; light-field microscopy; neural circuits; two-photon imaging |
| Substance Nomenclature: | SY7Q814VUP (Calcium) |
| Entry Date(s): | Date Created: 20251125 Date Completed: 20251125 Latest Revision: 20251125 |
| Update Code: | 20251126 |
| DOI: | 10.1073/pnas.2510337122 |
| PMID: | 41289378 |
| Datenbank: | MEDLINE |
Nájsť tento článok vo Web of Science | Abstract: | Competing Interests: Competing interests statement:The authors declare no competing interest.<br />Light-field microscopy (LFM) enables high-throughput functional imaging by scanlessly encoding entire volumes in single snapshots. However, LFM's computational burden and vulnerability to scattering limit its application to biological imaging. We present a light-field strategy for volumetric, scattering-mitigated neural circuit activity monitoring. A physics-based deep neural network, 2PiLnet, is trained with two-photon volumes and one-photon light fields. Light-field videos of jGCaMP8f-expressing neurons are acquired in neocortical brain slices. 2PiLnet reconstructs volumes with two-photon-like contrast and source confinement from scattered, blurry one-photon light fields from fields-of-view for which no two-photon images are provided. This enables automated segmentation and extraction of calcium signals with high signal-to-noise ratios and reduces optical crosstalk compared to conventional volume reconstruction methods. Imaging 100 volumes per second, we observe putative spikes fired at up to 10 Hz and the spatial intermingling of putative ensembles throughout [Formula: see text]-micron volumes. Compared to iterative algorithms, 2PiLnet workflows reduce light-field video processing times by several-fold, advancing the goal of real-time, scattering-robust volumetric neural circuit imaging for closed-loop and adaptive experimental paradigms. |
|---|---|
| ISSN: | 1091-6490 |
| DOI: | 10.1073/pnas.2510337122 |