Deep-Learning-Based Image Reconstruction and Enhancement in Optical Microscopy
Problem
Images from optical microscopes often need computational reconstruction or enhancement, and many inverse problems in microscopy are hard to solve with traditional methods.
Why it matters
Better microscopy imaging underpins biomedical research and diagnostics that hardware alone cannot deliver.
What the AI did
The article reviews deep neural networks that transform images between microscopy systems, add capabilities to existing techniques, and solve inverse problems from microscopy image data.
Results
not stated in source
Impact
Evidence supports only that deep learning is advancing computational microscopy toward blended instrument-compute designs.
Record
| Venue | Proceedings of the IEEE |
|---|---|
| Published | 2019 |
| Work type | article |
| Authors | Kevin d. Haan, Yair Rivenson, Yichen Wu, Aydogan Özcan |
| Citations | 136 (OpenAlex, at import time) |
| References | 195 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1109/jproc.2019.2949575 |
| Open access | hybrid |
| Keywords | deep learning, optical microscopy, image reconstruction, inverse problems, computational microscopy, biomedical imaging |
Importance rating
real but narrower advance
Computational microscopy underpins biomedical research and diagnostics, and deep learning is described as transformative for reconstruction/enhancement and blended instrument-compute designs, but this is a review with no quantified problem scale.