A probabilistic atlas for cell identification

Bubnis, Gregory; Ban, Steven; DiFranco, Matthew D.; Kato, Saul

doi:10.48550/arxiv.1903.09227

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cell matching can also be carried out between an imaging volume and a standard annotated atlas (e.g., OpenWorm [16] or NeuroPAL [17]) with known cell-type identities (see Digital ID, cell-type ID, and human annotation for a definition). Neurons or muscle cells in fixed and straightened worms can be recognized and mapped to the atlas using registration methods [18] and more recently, probabilistic models that take into account variability of cell positions [19][20][21] and take advantage of color information [17] to improve the recognition accuracy [20,21]. In this paper, we restrict ourselves to matching neuronal digital identities within an animal; matching cell-type identities across animals is beyond the scope of the present work.…”

Section: Introductionmentioning

confidence: 99%

Rapid detection and recognition of whole brain activity in a freely behaving Caenorhabditis elegans

Wu²,

Wang³

et al. 2022

PLoS Comput Biol

View full text Add to dashboard Cite

Advanced volumetric imaging methods and genetically encoded activity indicators have permitted a comprehensive characterization of whole brain activity at single neuron resolution in Caenorhabditis elegans. The constant motion and deformation of the nematode nervous system, however, impose a great challenge for consistent identification of densely packed neurons in a behaving animal. Here, we propose a cascade solution for long-term and rapid recognition of head ganglion neurons in a freely moving C. elegans. First, potential neuronal regions from a stack of fluorescence images are detected by a deep learning algorithm. Second, 2-dimensional neuronal regions are fused into 3-dimensional neuron entities. Third, by exploiting the neuronal density distribution surrounding a neuron and relative positional information between neurons, a multi-class artificial neural network transforms engineered neuronal feature vectors into digital neuronal identities. With a small number of training samples, our bottom-up approach is able to process each volume—1024 × 1024 × 18 in voxels—in less than 1 second and achieves an accuracy of 91% in neuronal detection and above 80% in neuronal tracking over a long video recording. Our work represents a step towards rapid and fully automated algorithms for decoding whole brain activity underlying naturalistic behaviors.

show abstract

Section: Introductionmentioning

confidence: 99%

Rapid detection and recognition of whole brain activity in a freely behaving Caenorhabditis elegans

Wu²,

Wang³

et al. 2022

PLoS Comput Biol

View full text Add to dashboard Cite

show abstract

Realization of cellomics to dive into the whole-body or whole-organ cell cloud

Mitani,

Susaki,

Matsumoto

et al. 2024

Nat Methods

View full text Add to dashboard Cite

Deep Neural Networks to Register and Annotate the Cells of theC. elegansNervous System

Atanas,

Lu,

Kim

et al. 2024

Preprint

View full text Add to dashboard Cite

Aligning and annotating the heterogeneous cell types that make up complex cellular tissues remains a major challenge in the analysis of biomedical imaging data. Here, we present a series of deep neural networks that allow for automatic non-rigid registration and cell identification in the context of the nervous system of freely-movingC. elegans. A semi-supervised learning approach was used to train aC. elegansregistration network (BrainAlignNet) that aligns pairs of images of the bendingC. eleganshead with single pixel-level accuracy. When incorporated into an image analysis pipeline, this network can link neuronal identities over time with 99.6% accuracy. A separate network (AutoCellLabeler) was trained to annotate >100 neuronal cell types in theC. eleganshead based on multi-spectral fluorescence of genetic markers. This network labels >100 different cell types per animal with 98% accuracy, exceeding individual human labeler performance by aggregating knowledge across manually labeled datasets. Finally, we trained a third network (CellDiscoveryNet) to perform unsupervised discovery and labeling of >100 cell types in theC. elegansnervous system by analyzing unlabeled multi-spectral imaging data from many animals. The performance of CellDiscoveryNet matched that of trained human labelers. These tools will be useful for a wide range of applications inC. elegansresearch and should be straightforward to generalize to many other applications requiring alignment and annotation of dense heterogeneous cell types in complex tissues.

show abstract

A probabilistic atlas for cell identification

Cited by 3 publications

References 14 publications

Rapid detection and recognition of whole brain activity in a freely behaving Caenorhabditis elegans

Rapid detection and recognition of whole brain activity in a freely behaving Caenorhabditis elegans

Realization of cellomics to dive into the whole-body or whole-organ cell cloud

Deep Neural Networks to Register and Annotate the Cells of theC. elegansNervous System

Contact Info

Product

Resources

About