Practical sensorless aberration estimation for 3D microscopy with deep learning
Estimation of optical aberrations from volumetric intensity images is a key step in sensorless adaptive optics for 3D microscopy. Recent approaches based on deep learning promise accurate results at fast processing speeds. However, collecting ground truth microscopy data for training the network is...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Optical Society of America
2020
|
| _version_ | 1797082047598034944 |
|---|---|
| author | Saha, D Schmidt, U Zhang, Q Barbotin, A Hu, Q Ji, N Booth, M Weigert, M Myers, E |
| author_facet | Saha, D Schmidt, U Zhang, Q Barbotin, A Hu, Q Ji, N Booth, M Weigert, M Myers, E |
| author_sort | Saha, D |
| collection | OXFORD |
| description | Estimation of optical aberrations from volumetric intensity images is a key step in sensorless adaptive optics for 3D microscopy. Recent approaches based on deep learning promise accurate results at fast processing speeds. However, collecting ground truth microscopy data for training the network is typically very difficult or even impossible thereby limiting this approach in practice. Here, we demonstrate that neural networks trained only on simulated data yield accurate predictions for real experimental images. We validate our approach on simulated and experimental datasets acquired with two different microscopy modalities and also compare the results to non-learned methods. Additionally, we study the predictability of individual aberrations with respect to their data requirements and find that the symmetry of the wavefront plays a crucial role. Finally, we make our implementation freely available as open source software in Python. |
| first_indexed | 2024-03-07T01:22:40Z |
| format | Journal article |
| id | oxford-uuid:90e42baf-c7f4-4c05-af6a-ca4ed4459183 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T01:22:40Z |
| publishDate | 2020 |
| publisher | Optical Society of America |
| record_format | dspace |
| spelling | oxford-uuid:90e42baf-c7f4-4c05-af6a-ca4ed44591832022-03-26T23:14:51ZPractical sensorless aberration estimation for 3D microscopy with deep learningJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:90e42baf-c7f4-4c05-af6a-ca4ed4459183EnglishSymplectic ElementsOptical Society of America2020Saha, DSchmidt, UZhang, QBarbotin, AHu, QJi, NBooth, MWeigert, MMyers, EEstimation of optical aberrations from volumetric intensity images is a key step in sensorless adaptive optics for 3D microscopy. Recent approaches based on deep learning promise accurate results at fast processing speeds. However, collecting ground truth microscopy data for training the network is typically very difficult or even impossible thereby limiting this approach in practice. Here, we demonstrate that neural networks trained only on simulated data yield accurate predictions for real experimental images. We validate our approach on simulated and experimental datasets acquired with two different microscopy modalities and also compare the results to non-learned methods. Additionally, we study the predictability of individual aberrations with respect to their data requirements and find that the symmetry of the wavefront plays a crucial role. Finally, we make our implementation freely available as open source software in Python. |
| spellingShingle | Saha, D Schmidt, U Zhang, Q Barbotin, A Hu, Q Ji, N Booth, M Weigert, M Myers, E Practical sensorless aberration estimation for 3D microscopy with deep learning |
| title | Practical sensorless aberration estimation for 3D microscopy with deep learning |
| title_full | Practical sensorless aberration estimation for 3D microscopy with deep learning |
| title_fullStr | Practical sensorless aberration estimation for 3D microscopy with deep learning |
| title_full_unstemmed | Practical sensorless aberration estimation for 3D microscopy with deep learning |
| title_short | Practical sensorless aberration estimation for 3D microscopy with deep learning |
| title_sort | practical sensorless aberration estimation for 3d microscopy with deep learning |
| work_keys_str_mv | AT sahad practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT schmidtu practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT zhangq practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT barbotina practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT huq practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT jin practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT boothm practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT weigertm practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning AT myerse practicalsensorlessaberrationestimationfor3dmicroscopywithdeeplearning |