Dataset Information

Segmentation of Neurons from Fluorescence Calcium Recordings Beyond Real-time.

ABSTRACT: Fluorescent genetically encoded calcium indicators and two-photon microscopy help understand brain function by generating large-scale in vivo recordings in multiple animal models. Automatic, fast, and accurate active neuron segmentation is critical when processing these videos. In this work, we developed and characterized a novel method, Shallow U-Net Neuron Segmentation (SUNS), to quickly and accurately segment active neurons from two-photon fluorescence imaging videos. We used temporal filtering and whitening schemes to extract temporal features associated with active neurons, and used a compact shallow U-Net to extract spatial features of neurons. Our method was both more accurate and an order of magnitude faster than state-of-the-art techniques when processing multiple datasets acquired by independent experimental groups; the difference in accuracy was enlarged when processing datasets containing few manually marked ground truths. We also developed an online version, potentially enabling real-time feedback neuroscience experiments.

SUBMITTER: Bao Y

PROVIDER: S-EPMC8415119 | biostudies-literature | 2021 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Segmentation of Neurons from Fluorescence Calcium Recordings Beyond Real-time.

Bao Yijun Y Soltanian-Zadeh Somayyeh S Farsiu Sina S Gong Yiyang Y

Nature machine intelligence 20210520 7

Fluorescent genetically encoded calcium indicators and two-photon microscopy help understand brain function by generating large-scale <i>in vivo</i> recordings in multiple animal models. Automatic, fast, and accurate active neuron segmentation is critical when processing these videos. In this work, we developed and characterized a novel method, Shallow U-Net Neuron Segmentation (SUNS), to quickly and accurately segment active neurons from two-photon fluorescence imaging videos. We used temporal ...[more]

PMID: 34485824

Similar Datasets

Project description:Osteoclasts are multinucleated cells of hematopoietic origin which are critically involved in physiological and pathological bone resorption. They develop from myeloid progenitors through characteristic gene expression changes and intercellular fusion. This process is directed by M-CSF and RANKL which are also able to trigger osteoclast development from bone marrow cells in vitro. Osteoclasts are conventionally visualized by histochemical staining followed by manual counting, which hinders kinetic studies and automated quantification. Here we describe two fluorescence-based assays for the real-time analysis of myeloid cell to osteoclast development (FRAMCO) in primary mouse bone marrow cell cultures. Both assays rely on red-to-green fluorescence conversion of the membrane-targeted tdTomato/membrane-targeted eGFP (mTmG) transgene by Cre recombinase driven by the osteoclast-specific cathepsin K promoter (Ctsk-Cre). In the first assay (FRAMCO1.1), osteoclast-specific gene expression triggers red-to-green color conversion of cells carrying both the Ctsk-Cre and mTmG transgenes. In the second assay (FRAMCO1.2), red-to-green fluorescence conversion is triggered by fusion of neighboring co-cultured bone marrow cells separately carrying either the Ctsk-Cre or the mTmG transgenes. The two assays were tested using a high-content confocal fluorescence imaging system, followed by automated quantification. The FRAMCO1.1 assay showed robust red-to-green fluorescence conversion of more than 50% of the culture (including mononuclear cells) within 3 days under osteoclastogenic conditions. The FRAMCO1.2 assay showed a less robust but still readily measurable red-to-green color conversion in multinuclear cells within 5 days of differentiation. The assays required both the Ctsk-Cre and the mTmG transgenes and gave no signals in parallel macrophage cultures. The proper functioning of the two assays was also confirmed at the DNA, mRNA and bulk protein level. The assay systems were validated using lisophosphatidylcholine, a previously reported inhibitor of preosteoclast fusion. Taken together, our assays allow high-throughput automated real-time analysis of two critical aspects of osteoclast development, facilitating the screening for novel drug candidates for the pharmacological control of osteoclast-mediated bone resorption.

Dataset Information

Segmentation of Neurons from Fluorescence Calcium Recordings Beyond Real-time.

Publications

Segmentation of Neurons from Fluorescence Calcium Recordings Beyond Real-time.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets