Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural Network
As a supplement to optical super-resolution microscopy techniques, computational super-resolution methods have demonstrated remarkable results in alleviating the spatiotemporal imaging trade-off. However, they commonly suffer from low structural fidelity and universality. Therefore, we herein propos...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Association for the Advancement of Science (AAAS)
2024-01-01
|
| Series: | Intelligent Computing |
| Online Access: | https://spj.science.org/doi/10.34133/icomputing.0082 |
| _version_ | 1827341174656991232 |
|---|---|
| author | Zitong Ye Yuran Huang Jinfeng Zhang Yunbo Chen Hanchu Ye Cheng Ji Luhong Jin Yanhong Gan Yile Sun Wenli Tao Yubing Han Xu Liu Youhua Chen Cuifang Kuang Wenjie Liu |
| author_facet | Zitong Ye Yuran Huang Jinfeng Zhang Yunbo Chen Hanchu Ye Cheng Ji Luhong Jin Yanhong Gan Yile Sun Wenli Tao Yubing Han Xu Liu Youhua Chen Cuifang Kuang Wenjie Liu |
| author_sort | Zitong Ye |
| collection | DOAJ |
| description | As a supplement to optical super-resolution microscopy techniques, computational super-resolution methods have demonstrated remarkable results in alleviating the spatiotemporal imaging trade-off. However, they commonly suffer from low structural fidelity and universality. Therefore, we herein propose a deep-physics-informed sparsity framework designed holistically to synergize the strengths of physical imaging models (image blurring processes), prior knowledge (continuity and sparsity constraints), a back-end optimization algorithm (image deblurring), and deep learning (an unsupervised neural network). Owing to the utilization of a multipronged learning strategy, the trained network can be applied to a variety of imaging modalities and samples to enhance the physical resolution by a factor of at least 1.67 without requiring additional training or parameter tuning. Given the advantages of high accessibility and universality, the proposed deep-physics-informed sparsity method will considerably enhance existing optical and computational imaging techniques and have a wide range of applications in biomedical research. |
| first_indexed | 2024-03-07T21:37:00Z |
| format | Article |
| id | doaj.art-d4c968266cdf43958ad4c71127054bea |
| institution | Directory Open Access Journal |
| issn | 2771-5892 |
| language | English |
| last_indexed | 2024-03-07T21:37:00Z |
| publishDate | 2024-01-01 |
| publisher | American Association for the Advancement of Science (AAAS) |
| record_format | Article |
| series | Intelligent Computing |
| spelling | doaj.art-d4c968266cdf43958ad4c71127054bea2024-02-26T13:16:39ZengAmerican Association for the Advancement of Science (AAAS)Intelligent Computing2771-58922024-01-01310.34133/icomputing.0082Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural NetworkZitong Ye0Yuran Huang1Jinfeng Zhang2Yunbo Chen3Hanchu Ye4Cheng Ji5Luhong Jin6Yanhong Gan7Yile Sun8Wenli Tao9Yubing Han10Xu Liu11Youhua Chen12Cuifang Kuang13Wenjie Liu14State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.Key Laboratory for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China.As a supplement to optical super-resolution microscopy techniques, computational super-resolution methods have demonstrated remarkable results in alleviating the spatiotemporal imaging trade-off. However, they commonly suffer from low structural fidelity and universality. Therefore, we herein propose a deep-physics-informed sparsity framework designed holistically to synergize the strengths of physical imaging models (image blurring processes), prior knowledge (continuity and sparsity constraints), a back-end optimization algorithm (image deblurring), and deep learning (an unsupervised neural network). Owing to the utilization of a multipronged learning strategy, the trained network can be applied to a variety of imaging modalities and samples to enhance the physical resolution by a factor of at least 1.67 without requiring additional training or parameter tuning. Given the advantages of high accessibility and universality, the proposed deep-physics-informed sparsity method will considerably enhance existing optical and computational imaging techniques and have a wide range of applications in biomedical research.https://spj.science.org/doi/10.34133/icomputing.0082 |
| spellingShingle | Zitong Ye Yuran Huang Jinfeng Zhang Yunbo Chen Hanchu Ye Cheng Ji Luhong Jin Yanhong Gan Yile Sun Wenli Tao Yubing Han Xu Liu Youhua Chen Cuifang Kuang Wenjie Liu Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural Network Intelligent Computing |
| title | Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural Network |
| title_full | Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural Network |
| title_fullStr | Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural Network |
| title_full_unstemmed | Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural Network |
| title_short | Universal and High-Fidelity Resolution Extending for Fluorescence Microscopy Using a Single-Training Physics-Informed Sparse Neural Network |
| title_sort | universal and high fidelity resolution extending for fluorescence microscopy using a single training physics informed sparse neural network |
| url | https://spj.science.org/doi/10.34133/icomputing.0082 |
| work_keys_str_mv | AT zitongye universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT yuranhuang universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT jinfengzhang universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT yunbochen universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT hanchuye universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT chengji universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT luhongjin universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT yanhonggan universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT yilesun universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT wenlitao universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT yubinghan universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT xuliu universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT youhuachen universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT cuifangkuang universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork AT wenjieliu universalandhighfidelityresolutionextendingforfluorescencemicroscopyusingasingletrainingphysicsinformedsparseneuralnetwork |