Approaching maximum resolution in structured illumination microscopy via accurate noise modeling
Abstract Biological images captured by microscopes are characterized by heterogeneous signal-to-noise ratios (SNRs) due to spatially varying photon emission across the field of view convoluted with camera noise. State-of-the-art unsupervised structured illumination microscopy (SIM) reconstruction me...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | npj Imaging |
| Online Access: | https://doi.org/10.1038/s44303-024-00066-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571468783812608 |
|---|---|
| author | Ayush Saurabh Peter T. Brown J. Shepard Bryan IV Zachary R. Fox Rory Kruithoff Cristopher Thompson Comert Kural Douglas P. Shepherd Steve Pressé |
| author_facet | Ayush Saurabh Peter T. Brown J. Shepard Bryan IV Zachary R. Fox Rory Kruithoff Cristopher Thompson Comert Kural Douglas P. Shepherd Steve Pressé |
| author_sort | Ayush Saurabh |
| collection | DOAJ |
| description | Abstract Biological images captured by microscopes are characterized by heterogeneous signal-to-noise ratios (SNRs) due to spatially varying photon emission across the field of view convoluted with camera noise. State-of-the-art unsupervised structured illumination microscopy (SIM) reconstruction methods, commonly implemented in the Fourier domain, often do not accurately model this noise. Such methods therefore suffer from high-frequency artifacts, user-dependent choices of smoothness constraints making assumptions on biological features, and unphysical negative values in the recovered fluorescence intensity map. On the other hand, supervised algorithms rely on large datasets for training, and often require retraining for new sample structures. Consequently, achieving high contrast near the maximum theoretical resolution in an unsupervised, physically principled manner remains an open problem. Here, we propose Bayesian-SIM (B-SIM), a Bayesian framework to quantitatively reconstruct SIM data, rectifying these shortcomings by accurately incorporating known noise sources in the spatial domain. To accelerate the reconstruction process, we use the finite extent of the point-spread-function to devise a parallelized Monte Carlo strategy involving chunking and restitching of the inferred fluorescence intensity. We benchmark our framework on both simulated and experimental images, and demonstrate improved contrast permitting feature recovery at up to 25% shorter length scales over state-of-the-art methods at both high- and low SNR. B-SIM enables unsupervised, quantitative, physically accurate reconstruction without the need for labeled training data, democratizing high-quality SIM reconstruction and expands the capabilities of live-cell SIM to lower SNR, potentially revealing biological features in previously inaccessible regimes. |
| format | Article |
| id | doaj-art-abb645bb2bdb48c1aefbc528e4e910ea |
| institution | Kabale University |
| issn | 2948-197X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Imaging |
| spelling | doaj-art-abb645bb2bdb48c1aefbc528e4e910ea2025-02-02T12:37:07ZengNature Portfolionpj Imaging2948-197X2025-01-013111410.1038/s44303-024-00066-8Approaching maximum resolution in structured illumination microscopy via accurate noise modelingAyush Saurabh0Peter T. Brown1J. Shepard Bryan IV2Zachary R. Fox3Rory Kruithoff4Cristopher Thompson5Comert Kural6Douglas P. Shepherd7Steve Pressé8Center for Biological Physics, Arizona State UniversityCenter for Biological Physics, Arizona State UniversityCenter for Biological Physics, Arizona State UniversityComputational Science and Engineering Division, Oak Ridge National LaboratoryCenter for Biological Physics, Arizona State UniversityDepartment of Physics, The Ohio State UniversityDepartment of Physics, The Ohio State UniversityCenter for Biological Physics, Arizona State UniversityCenter for Biological Physics, Arizona State UniversityAbstract Biological images captured by microscopes are characterized by heterogeneous signal-to-noise ratios (SNRs) due to spatially varying photon emission across the field of view convoluted with camera noise. State-of-the-art unsupervised structured illumination microscopy (SIM) reconstruction methods, commonly implemented in the Fourier domain, often do not accurately model this noise. Such methods therefore suffer from high-frequency artifacts, user-dependent choices of smoothness constraints making assumptions on biological features, and unphysical negative values in the recovered fluorescence intensity map. On the other hand, supervised algorithms rely on large datasets for training, and often require retraining for new sample structures. Consequently, achieving high contrast near the maximum theoretical resolution in an unsupervised, physically principled manner remains an open problem. Here, we propose Bayesian-SIM (B-SIM), a Bayesian framework to quantitatively reconstruct SIM data, rectifying these shortcomings by accurately incorporating known noise sources in the spatial domain. To accelerate the reconstruction process, we use the finite extent of the point-spread-function to devise a parallelized Monte Carlo strategy involving chunking and restitching of the inferred fluorescence intensity. We benchmark our framework on both simulated and experimental images, and demonstrate improved contrast permitting feature recovery at up to 25% shorter length scales over state-of-the-art methods at both high- and low SNR. B-SIM enables unsupervised, quantitative, physically accurate reconstruction without the need for labeled training data, democratizing high-quality SIM reconstruction and expands the capabilities of live-cell SIM to lower SNR, potentially revealing biological features in previously inaccessible regimes.https://doi.org/10.1038/s44303-024-00066-8 |
| spellingShingle | Ayush Saurabh Peter T. Brown J. Shepard Bryan IV Zachary R. Fox Rory Kruithoff Cristopher Thompson Comert Kural Douglas P. Shepherd Steve Pressé Approaching maximum resolution in structured illumination microscopy via accurate noise modeling npj Imaging |
| title | Approaching maximum resolution in structured illumination microscopy via accurate noise modeling |
| title_full | Approaching maximum resolution in structured illumination microscopy via accurate noise modeling |
| title_fullStr | Approaching maximum resolution in structured illumination microscopy via accurate noise modeling |
| title_full_unstemmed | Approaching maximum resolution in structured illumination microscopy via accurate noise modeling |
| title_short | Approaching maximum resolution in structured illumination microscopy via accurate noise modeling |
| title_sort | approaching maximum resolution in structured illumination microscopy via accurate noise modeling |
| url | https://doi.org/10.1038/s44303-024-00066-8 |
| work_keys_str_mv | AT ayushsaurabh approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT petertbrown approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT jshepardbryaniv approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT zacharyrfox approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT rorykruithoff approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT cristopherthompson approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT comertkural approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT douglaspshepherd approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling AT stevepresse approachingmaximumresolutioninstructuredilluminationmicroscopyviaaccuratenoisemodeling |