Results from Cell-TRACTR and model overview for object detection.

<p>(A) Representative output from Cell-TRACTR for a single chamber from a time-lapse microscopy movie of <i>E. coli</i> bacteria growing in the mother machine microfluidic device. The output is displayed as a kymograph where the same chamber is shown over time. Each color represent...

Full description

Saved in:
Bibliographic Details
Main Author: Owen M. O’Connor (11965460) (author)
Other Authors: Mary J. Dunlop (463579) (author)
Published: 2025
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1852019910835699712
author Owen M. O’Connor (11965460)
author2 Mary J. Dunlop (463579)
author2_role author
author_facet Owen M. O’Connor (11965460)
Mary J. Dunlop (463579)
author_role author
dc.creator.none.fl_str_mv Owen M. O’Connor (11965460)
Mary J. Dunlop (463579)
dc.date.none.fl_str_mv 2025-05-29T12:19:51Z
dc.identifier.none.fl_str_mv 10.1371/journal.pcbi.1013071.g001
dc.relation.none.fl_str_mv https://figshare.com/articles/figure/Results_from_Cell-TRACTR_and_model_overview_for_object_detection_/29181965
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biophysics
Genetics
Cancer
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
track many objects
present unique challenges
global contextual dependencies
existing architectures developed
easily interpretable assessment
defined microfluidic geometry
convolutional neural networks
bacteria growing within
div >< p
tr </ u
r </ u
c </ u
division accuracy compared
tracking cells without
assess cell division
cells </ p
tractr (< u
cell tracking based
</ u
division accuracy
tracking cells
cell tracking
based models
based architecture
cell segmentation
work establishes
two dimensions
tractr operates
simultaneously segmenting
results demonstrate
new framework
medical data
fall short
ecognition ),
deep learning
data analysis
art algorithms
analyzing biological
dc.title.none.fl_str_mv Results from Cell-TRACTR and model overview for object detection.
dc.type.none.fl_str_mv Image
Figure
info:eu-repo/semantics/publishedVersion
image
description <p>(A) Representative output from Cell-TRACTR for a single chamber from a time-lapse microscopy movie of <i>E. coli</i> bacteria growing in the mother machine microfluidic device. The output is displayed as a kymograph where the same chamber is shown over time. Each color represents a unique cell being tracked and black arrows signify cell divisions. This movie is from the test set. (B) Overview of Cell-TRACTR tracking mammalian nuclei across multiple frames. Each color represents a unique cell being tracked. These frames are from the DeepCell dataset using data from the test set. (C) Cell-TRACTR is a DETR-based model that first uses a CNN backbone to extract image features, then uses a transformer encoder-decoder framework that leverages attention mechanisms. Encoded multi-scale features are used for query selection to initialize the object queries. Object queries attend to the encoded multi-scale features to generate the output embeddings needed to predict class labels, bounding boxes, and segmentation masks. If the model predicts an object query to be a cell, then the output embedding is shown as a colored box in the schematic. Object queries predicted to be of class “no object” are colored black and are discarded. The squares represent content embeddings and the circles represent positional embeddings.</p>
eu_rights_str_mv openAccess
id Manara_dcaa81a1dedfa65a897cdb054f2e8002
identifier_str_mv 10.1371/journal.pcbi.1013071.g001
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/29181965
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Results from Cell-TRACTR and model overview for object detection.Owen M. O’Connor (11965460)Mary J. Dunlop (463579)BiophysicsGeneticsCancerBiological Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedtrack many objectspresent unique challengesglobal contextual dependenciesexisting architectures developedeasily interpretable assessmentdefined microfluidic geometryconvolutional neural networksbacteria growing withindiv >< ptr </ ur </ uc </ udivision accuracy comparedtracking cells withoutassess cell divisioncells </ ptractr (< ucell tracking based</ udivision accuracytracking cellscell trackingbased modelsbased architecturecell segmentationwork establishestwo dimensionstractr operatessimultaneously segmentingresults demonstratenew frameworkmedical datafall shortecognition ),deep learningdata analysisart algorithmsanalyzing biological<p>(A) Representative output from Cell-TRACTR for a single chamber from a time-lapse microscopy movie of <i>E. coli</i> bacteria growing in the mother machine microfluidic device. The output is displayed as a kymograph where the same chamber is shown over time. Each color represents a unique cell being tracked and black arrows signify cell divisions. This movie is from the test set. (B) Overview of Cell-TRACTR tracking mammalian nuclei across multiple frames. Each color represents a unique cell being tracked. These frames are from the DeepCell dataset using data from the test set. (C) Cell-TRACTR is a DETR-based model that first uses a CNN backbone to extract image features, then uses a transformer encoder-decoder framework that leverages attention mechanisms. Encoded multi-scale features are used for query selection to initialize the object queries. Object queries attend to the encoded multi-scale features to generate the output embeddings needed to predict class labels, bounding boxes, and segmentation masks. If the model predicts an object query to be a cell, then the output embedding is shown as a colored box in the schematic. Object queries predicted to be of class “no object” are colored black and are discarded. The squares represent content embeddings and the circles represent positional embeddings.</p>2025-05-29T12:19:51ZImageFigureinfo:eu-repo/semantics/publishedVersionimage10.1371/journal.pcbi.1013071.g001https://figshare.com/articles/figure/Results_from_Cell-TRACTR_and_model_overview_for_object_detection_/29181965CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/291819652025-05-29T12:19:51Z
spellingShingle Results from Cell-TRACTR and model overview for object detection.
Owen M. O’Connor (11965460)
Biophysics
Genetics
Cancer
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
track many objects
present unique challenges
global contextual dependencies
existing architectures developed
easily interpretable assessment
defined microfluidic geometry
convolutional neural networks
bacteria growing within
div >< p
tr </ u
r </ u
c </ u
division accuracy compared
tracking cells without
assess cell division
cells </ p
tractr (< u
cell tracking based
</ u
division accuracy
tracking cells
cell tracking
based models
based architecture
cell segmentation
work establishes
two dimensions
tractr operates
simultaneously segmenting
results demonstrate
new framework
medical data
fall short
ecognition ),
deep learning
data analysis
art algorithms
analyzing biological
status_str publishedVersion
title Results from Cell-TRACTR and model overview for object detection.
title_full Results from Cell-TRACTR and model overview for object detection.
title_fullStr Results from Cell-TRACTR and model overview for object detection.
title_full_unstemmed Results from Cell-TRACTR and model overview for object detection.
title_short Results from Cell-TRACTR and model overview for object detection.
title_sort Results from Cell-TRACTR and model overview for object detection.
topic Biophysics
Genetics
Cancer
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
track many objects
present unique challenges
global contextual dependencies
existing architectures developed
easily interpretable assessment
defined microfluidic geometry
convolutional neural networks
bacteria growing within
div >< p
tr </ u
r </ u
c </ u
division accuracy compared
tracking cells without
assess cell division
cells </ p
tractr (< u
cell tracking based
</ u
division accuracy
tracking cells
cell tracking
based models
based architecture
cell segmentation
work establishes
two dimensions
tractr operates
simultaneously segmenting
results demonstrate
new framework
medical data
fall short
ecognition ),
deep learning
data analysis
art algorithms
analyzing biological