Statistically unbiased prediction enables accurate denoising of voltage imaging data.

Statistically unbiased prediction enables accurate denoising of voltage imaging data.
复制标题

统计上无偏见的预测可以准确地降低电压成像数据。

DOI:
10.1038/s41592-023-02005-8
复制
发表时间:
2023-10
期刊:
影响因子:
48
通讯作者:
Yoon, Young-Gyu
Yoon, Young-Gyu
中科院分区:
生物学1区
文献类型:
--
作者:
Eom, Minho;Han, Seungjae;Park, Pojeong;Kim, Gyuri;Cho, Eun-Seo;Sim, Jueun;Lee, Kang-Han;Kim, Seonghoon;Tian, He;Boehm, Urs L.;Lowet, Eric;Tseng, Hua-an;Choi, Jieun;Lucia, Stephani Edwina;Ryu, Seung Hyun;Rozsa, Marton;Chang, Sunghoe;Kim, Pilhan;Han, Xue;Piatkevich, Kiryl D.;Choi, Myunghwan;Kim, Cheol-Hee;Cohen, Adam E.;Chang, Jae-Byum;Yoon, Young-Gyu

文献摘要

参考文献

相似文献

在这里,我们报告了 SUPPORT(利用成像数据中的时空信息进行统计无偏预测),这是一种用于消除电压成像数据中的泊松高斯噪声的自监督学习方法。支持基于这样的见解:电压成像数据中的像素值高度依赖于其时空相邻像素,即使其时间相邻帧本身不能提供用于统计预测的有用信息。这种依赖性被具有时空盲点的卷积神经网络捕获并使用,以精确地对电压成像数据进行去噪,其中某个时间帧中动作电位的存在无法通过其他帧中的信息推断出来。通过模拟和实验,我们表明 SUPPORT 可以对电压成像数据和其他类型的显微图像进行精确去噪,同时保留场景内的潜在动态。利用成像数据中的时空信息进行统计无偏预测 (SUPPORT) 是一种自监督深度学习方法,可准确对电压和钙成像数据进行去噪,同时保留真实的动态信号。
Here we report SUPPORT (statistically unbiased prediction utilizing spatiotemporal information in imaging data), a self-supervised learning method for removing Poisson–Gaussian noise in voltage imaging data. SUPPORT is based on the insight that a pixel value in voltage imaging data is highly dependent on its spatiotemporal neighboring pixels, even when its temporally adjacent frames alone do not provide useful information for statistical prediction. Such dependency is captured and used by a convolutional neural network with a spatiotemporal blind spot to accurately denoise voltage imaging data in which the existence of the action potential in a time frame cannot be inferred by the information in other frames. Through simulations and experiments, we show that SUPPORT enables precise denoising of voltage imaging data and other types of microscopy image while preserving the underlying dynamics within the scene. Statistically unbiased prediction utilizing spatiotemporal information in imaging data (SUPPORT) is a self-supervised deep learning approach to accurately denoise voltage and calcium imaging data while preserving true dynamic signals.
DOI: 10.7554/elife.76993
发表时间: 2022-03-23
期刊: eLife
影响因子: 7.7
作者:
Landau AT;Park P;Wong-Campos JD;Tian H;Cohen AE;Sabatini BL
通讯作者: Sabatini BL
DOI: 10.1364/boe.485187
发表时间: 2023-04-01
影响因子: 3.4
作者:
Choi, Jieun;Choi, Min -sun;Kim, Pilhan
通讯作者: Kim, Pilhan
DOI: 10.1021/acsnano.8b02330
发表时间: 2018-07-01
期刊: ACS NANO
影响因子: 17.1
作者:
Kim, Jin Yong;Ahn, Jinhyo;Jon, Sangyong
通讯作者: Jon, Sangyong
DOI: 10.1016/j.neuron.2023.03.009
发表时间: 2023-05-17
期刊: NEURON
影响因子: 16.2
作者:
Abdelfattah, Ahmed S.;Zheng, Jihong;Kolb, Ilya
通讯作者: Kolb, Ilya
DOI: 10.1038/s41592-021-01285-2
发表时间: 2021-11
期刊: Nature methods
影响因子: 48
作者:
Lecoq J;Oliver M;Siegle JH;Orlova N;Ledochowitsch P;Koch C
通讯作者: Koch C