Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction
Abstract By circumventing the resolution limitations of optics, coherent diffractive imaging (CDI) and ptychography are making their way into scientific fields ranging from X-ray imaging to astronomy. Yet, the need for time consuming iterative phase recovery hampers real-time imaging. While supervis...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-12-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-48351-7 |
| _version_ | 1827399531213357056 |
|---|---|
| author | Oliver Hoidn Aashwin Ananda Mishra Apurva Mehta |
| author_facet | Oliver Hoidn Aashwin Ananda Mishra Apurva Mehta |
| author_sort | Oliver Hoidn |
| collection | DOAJ |
| description | Abstract By circumventing the resolution limitations of optics, coherent diffractive imaging (CDI) and ptychography are making their way into scientific fields ranging from X-ray imaging to astronomy. Yet, the need for time consuming iterative phase recovery hampers real-time imaging. While supervised deep learning strategies have increased reconstruction speed, they sacrifice image quality. Furthermore, these methods’ demand for extensive labeled training data is experimentally burdensome. Here, we propose an unsupervised physics-informed neural network reconstruction method, PtychoPINN, that retains the factor of 100-to-1000 speedup of deep learning-based reconstruction while improving reconstruction quality by combining the diffraction forward map with real-space constraints from overlapping measurements. In particular, PtychoPINN gains a factor of 4 in linear resolution and an 8 dB improvement in PSNR while also accruing improvements in generalizability and robustness. This blend of performance and computational efficiency offers exciting prospects for high-resolution real-time imaging in high-throughput environments such as X-ray free electron lasers (XFELs) and diffraction-limited light sources. |
| first_indexed | 2024-03-08T19:47:23Z |
| format | Article |
| id | doaj.art-6db33227cdc646a98b8f2c058c5a9929 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-08T19:47:23Z |
| publishDate | 2023-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-6db33227cdc646a98b8f2c058c5a99292023-12-24T12:18:24ZengNature PortfolioScientific Reports2045-23222023-12-0113111110.1038/s41598-023-48351-7Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstructionOliver Hoidn0Aashwin Ananda Mishra1Apurva Mehta2SLAC National Accelerator LaboratorySLAC National Accelerator LaboratorySLAC National Accelerator LaboratoryAbstract By circumventing the resolution limitations of optics, coherent diffractive imaging (CDI) and ptychography are making their way into scientific fields ranging from X-ray imaging to astronomy. Yet, the need for time consuming iterative phase recovery hampers real-time imaging. While supervised deep learning strategies have increased reconstruction speed, they sacrifice image quality. Furthermore, these methods’ demand for extensive labeled training data is experimentally burdensome. Here, we propose an unsupervised physics-informed neural network reconstruction method, PtychoPINN, that retains the factor of 100-to-1000 speedup of deep learning-based reconstruction while improving reconstruction quality by combining the diffraction forward map with real-space constraints from overlapping measurements. In particular, PtychoPINN gains a factor of 4 in linear resolution and an 8 dB improvement in PSNR while also accruing improvements in generalizability and robustness. This blend of performance and computational efficiency offers exciting prospects for high-resolution real-time imaging in high-throughput environments such as X-ray free electron lasers (XFELs) and diffraction-limited light sources.https://doi.org/10.1038/s41598-023-48351-7 |
| spellingShingle | Oliver Hoidn Aashwin Ananda Mishra Apurva Mehta Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction Scientific Reports |
| title | Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction |
| title_full | Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction |
| title_fullStr | Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction |
| title_full_unstemmed | Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction |
| title_short | Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction |
| title_sort | physics constrained unsupervised deep learning for rapid high resolution scanning coherent diffraction reconstruction |
| url | https://doi.org/10.1038/s41598-023-48351-7 |
| work_keys_str_mv | AT oliverhoidn physicsconstrainedunsuperviseddeeplearningforrapidhighresolutionscanningcoherentdiffractionreconstruction AT aashwinanandamishra physicsconstrainedunsuperviseddeeplearningforrapidhighresolutionscanningcoherentdiffractionreconstruction AT apurvamehta physicsconstrainedunsuperviseddeeplearningforrapidhighresolutionscanningcoherentdiffractionreconstruction |