Deep learning enhanced light sheet fluorescence microscopy for in vivo 4D imaging of zebrafish heart beating
Problem
Time-resolved volumetric fluorescence imaging over extended durations is limited by the trade-off among imaging speed, light exposure, illumination power and image quality.
Why it matters
Long-duration in vivo 4D imaging of living organisms, such as beating zebrafish hearts, is a key driver of biomedical research.
What the AI did
Developed UI-Trans, a CNN-transformer network, that restores rapid volumetric time-lapse light sheet microscopy images degraded by coupled noise and scattering, outperforming state-of-the-art deep learning networks.
Results
Restoration achieved with less than 0.03% light exposure and 3.3% acquisition time versus standard acquisition; outperforms state-of-the-art deep learning networks
Impact
Demonstrated in vivo 4D imaging of zebrafish heart beating with drastically reduced light exposure and acquisition time.
Record
| Venue | Light Science & Applications |
|---|---|
| Published | 2025 |
| Work type | article |
| Authors | Meng Zhang, Renjian Li, Songnian Fu, Sunil Kumar, James McGinty, Yuwen Qin, Lingling Chen |
| Citations | 22 (OpenAlex, at import time) |
| References | 59 |
| Model or method | UI-Trans (U-net integrated transformer) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41377-024-01710-z |
| Open access | gold |
| Keywords | light sheet microscopy, deep learning, zebrafish heart, 4D imaging, image restoration, transformer |
Importance rating
significant — thousands to millions affected
Abstract calls long-duration 4D in vivo imaging a key driver of biomedical research; demonstrated on zebrafish hearts; no human/industry scale given.