Deep learning-driven adaptive optics for single-molecule localization microscopy
Problem
Tissue-induced aberrations blur and distort single-molecule emission patterns, limiting resolution in single-molecule localization microscopy, and conventional sensorless adaptive optics is slow and inconsistent.
Why it matters
Aberration correction determines whether super-resolution imaging can resolve fine biological structure in tissue.
What the AI did
A trained deep neural network monitors individual emission patterns, directly infers shared wavefront distortion, and drives a deformable mirror for near real-time compensation, bypassing iterative trial-and-evaluate loops.
Results
Near real-time wavefront compensation demonstrated; quantitative results truncated in abstract.
Impact
Evidence supports the claimed near real-time aberration correction; independent benchmarking not stated.
Record
| Venue | Nature Methods |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Peiyi Zhang, Donghan Ma, Xi Cheng, Andy Po‐Yi Tsai, Yu Jie Tang, Hao-Cheng Gao, Li Fang, Cheng Bi et al. |
| Citations | 40 (OpenAlex, at import time) |
| References | 85 |
| Model or method | deep neural network (architecture not specified) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41592-023-02029-0 |
| Open access | hybrid |
| Keywords | adaptive optics, deep learning, SMLM, aberration correction, deformable mirror |
Importance rating
real but narrower advance
Aberration correction determines whether SMLM can resolve fine structure in tissue; near real-time AO is a new capability for super-resolution imaging, but no impact scale is stated.