| 时间 |
2022年9月6日，12:00-13:00 |
| 地点 |
生物医学馆E203 |
| 主持人 |
|
| 报告人 |
李欣阳 |
| 题目 |
Deep self-supervised denoising enables ultrasensitive fluorescence imaging beyond the shot-noise limit. |
| 摘要 | A fundamental challenge in fluorescence microscopy is the photon shot noise arising from the inevitable stochasticity of photon detection. Noise increases measurement uncertainty, degrades image quality, and limits imaging resolution, speed, and sensitivity. To achieve ultrasensitive imaging beyond the shot-noise limit, we developed DeepCAD, a self-supervised deep-learning method for noise suppression of fluorescence time-lapse imaging. DeepCAD can learn to denoise without seeing any high signal-to-noise ratio (SNR) observations. A high imaging SNR can be acquired with 10-fold fewer fluorescence photons. Recent advancements of DeepCAD allow real-time denoising on a two-photon microscope. We demonstrate the utility of DeepCAD in a series of photon-limited experiments, including in vivo calcium imaging of mouse, zebrafish larva, and fruit fly, recording of 3D migration of neutrophils, and imaging of 3D dynamics of cortical ATP release. DeepCAD will facilitate the morphological and functional interrogation of biological dynamics with a minimal photon budget. |
请扫二维码填写报名问卷（报告环节提供午餐）
