Abstract:Calcium imaging is susceptible to motion distortions and background noises, particularly for monitoring active animals under low-dose laser irradiation, and hence unavoidably hinder the critical analysis of neural functions. Current research efforts tend to focus on either denoising or dewarping and do not provide effective methods for videos distorted by both noises and motion artifacts simultaneously. We found that when a self-supervised denoising model of DeepCAD [Nat. Methods18, 1359 (2021)10.1038/s41592-021-01225-0] is used on the calcium imaging contaminated by noise and motion warping, it can remove the motion artifacts effectively but with regenerated noises. To address this issue, we develop a two-level deep-learning (DL) pipeline to dewarp and denoise the calcium imaging video sequentially. The pipeline consists of two 3D self-supervised DL models that do not require warp-free and high signal-to-noise ratio (SNR) observations for network optimization. Specifically, a high-frequency enhancement block is presented in the denoising network to restore more structure information in the denoising process; a hierarchical perception module and a multi-scale attention module are designed in the dewarping network to tackle distortions of various sizes. Experiments conducted on seven videos from two-photon and confocal imaging systems demonstrate that our two-level DL pipeline can restore high-clarity neuron images distorted by both motion warping and background noises. Compared to typical DeepCAD, our denoising model achieves a significant improvement of approximately 30% in image resolution and up to 28% in signal-to-noise ratio; compared to traditional dewarping and denoising methods, our proposed pipeline network recovers more neurons, enhancing signal fidelity and improving data correlation among frames by 35% and 60% respectively. This work may provide an attractive method for long-term neural activity monitoring in awake animals and also facilitate functional analysis of neural circuits.

An end-to-end recurrent compressed sensing method to denoise, detect and demix calcium imaging data

Reinforcing Neuron Extraction and Spike Inference in Calcium Imaging Using Deep Self-Supervised Denoising

Fast and robust active neuron segmentation in two-photon calcium imaging using spatiotemporal deep learning

Fast, efficient, and accurate neuro-imaging denoising via supervised deep learning

Penalized matrix decomposition for denoising, compression, and improved demixing of functional imaging data

Classification of A Pressing Movement by Background Removed Single- photon Calcium Image Sequences

Fast, Simple Calcium Imaging Segmentation with Fully Convolutional Networks

Simultaneous removal of noise and correction of motion warping in neuron calcium imaging using a pipeline structure of self-supervised deep learning models

Efficient and accurate extraction of in vivo calcium signals from microendoscopic video data

Rapid deep widefield neuron finder driven by virtual calcium imaging data

An Open-Source Real-Time Motion Correction Plug-In for Single-Photon Calcium Imaging of Head-Mounted Microscopy

Rapid detection of neurons in widefield calcium imaging datasets after training with synthetic data

Automatic Neuron Detection in Calcium Imaging Data Using Convolutional Networks

Decomposition-Estimation-Reconstruction: An Automatic and Accurate Neuron Extraction Paradigm

CalciSeg: A versatile approach for unsupervised segmentation of calcium imaging data

TAG-SPARK: Empowering High-Speed Volumetric Imaging With Deep Learning and Spatial Redundancy

Convolutional Neural Network Based Computational Imaging Spectroscopy

Neurons Identification of Single-Photon Wide-Field Calcium Fluorescent Imaging Data

Fast, scalable, and statistically robust cell extraction from large-scale neural calcium imaging datasets

A preprocessing toolbox for 2-photon subcellular calcium imaging

Reinforcing Neuron Extraction from Calcium Imaging Data Via Depth-Estimation Constrained Nonnegative Matrix Factorization