Deep learning for robust and flexible tracking in behavioral studies for C. elegans

Bates, Kathleen; Le, Kim; Lu, Hang

doi:10.1371/journal.pcbi.1009942

Cited by 22 publications

(23 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous study by K Bates et al . [39] used a Faster-RCNN based detection model solely without a tracking backbone for worm tracking. Also, the detection model alone cannot account for long term worm occlusion while tracking multiple worms.…”

Section: Resultsmentioning

confidence: 99%

Deep-Worm-Tracker: Deep Learning Methods for Accurate Detection and Tracking for Behavioral Studies in C. elegans

Banerjee

Khan

Sharma

2022

Preprint

View full text Add to dashboard Cite

Accurate detection and tracking of model organisms such as C. elegans worms remains a fundamental task in behavioral studies. Traditional Machine Learning (ML) and Computer Vision (CV) methods produce poor detection results and suffer from repeated ID switches during tracking under occlusions and noisy backgrounds. Using Deep Learning (DL) methods, the task of animal tracking from video recordings, like those in camera trap experiments, has become much more viable. The large amount of data generated in ethological studies, makes such models suitable for real world scenarios in the wild. We propose Deep-Worm-Tracker, an end to end DL model, which is a combination of You Only Look Once (YOLOv5) object detection model and Strong Simple Online Real Time Tracking (Strong SORT) tracking backbone that is highly accurate and provides tracking results in real time inference speeds. Present literature has few solutions to track animals under occlusions and even fewer publicly available large scale animal re-ID datasets. Thus, we also provide a worm re-ID dataset to minimize worm ID switches, which, to the best of our knowledge, is first-of-its-kind for C. elegans. We are able to track worms at a mean Average Precision (mAP@0.5) > 98% within just 9 minutes of training time with inference speeds of 9-15 ms for worm detection and on average 27 ms for worm tracking. Our tracking results show that Deep-Worm-Tracker is well suited for ethological studies involving C. elegans.

show abstract

Section: Resultsmentioning

confidence: 99%

Deep-Worm-Tracker: Deep Learning Methods for Accurate Detection and Tracking for Behavioral Studies in C. elegans

Banerjee

Khan

Sharma

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The deep-learning-aided machine vision methods we developed here, capable of segmenting individual animals and assaying them across different attributes, may find other applications for C. elegans studies, for example in analyses of locomotion 43,44 , aging 45,46 , and sleep 47 .…”

Section: Discussionmentioning

confidence: 99%

A robotic system for automated genetic manipulation and analysis ofCaenorhabditis elegans

Li¹,

Fouad²,

Bowlin³

et al. 2022

Preprint

View full text Add to dashboard Cite

The nematode C. elegans is one of the most widely studied organisms in biology due to its small size, rapid life cycle, and easily manipulated genetics. Research with C. elegans depends on labor-intensive and time-consuming manual procedures. Here we describe WormPicker, a robotic system capable of performing complex genetic manipulations and other tasks by imaging, phenotyping, and selectively transferring C. elegans on standard agar media. Our system uses a motorized stage to move an imaging system and a robotic arm over an array of plates. Machine vision algorithms identify worms and assay their phenotypes, and the robotic arm transfers worms between plates using an electrically self-sterilized wire loop. We developed software to enable the system to autonomously carry out complex protocols. We validated the effectiveness and versatility of our methods by automating a collection of common C. elegans procedures, including genetic crossing, genetic mapping, and genomic integration of a transgene.

show abstract

“…Recently, deep learning techniques have been utilized to track C. elegans worms using e.g. bounding box predictions [89][90][91] and fully resolved centre-line splines in the case of isolated worms [92], allowing for detection also during periods of self-overlap.…”

Section: Introductionmentioning

confidence: 99%

Fast spline detection in high density microscopy data

Alonso¹,

Kirkegaard²

2023

Preprint

View full text Add to dashboard Cite

Computer-aided analysis of biological microscopy data has seen a massive improvement with the utilization of general-purpose deep learning techniques. Yet, in microscopy studies of multi-organism systems, the problem of collision and overlap remains challenging. This is particularly true for systems composed of slender bodies such as crawling nematodes, swimming spermatozoa, or the beating of eukaryotic or prokaryotic flagella. Here, we develop a novel end-to-end deep learning approach to extract precise shape trajectories of generally motile and overlapping splines. Our method works in low resolution settings where feature keypoints are hard to define and detect. Detection is fast and we demonstrate the ability to track thousands of overlapping organisms simultaneously. While our approach is agnostic to area of application, we present it in the setting of and exemplify its usability on dense experiments of crawling Caenorhabditis elegans. The model training is achieved purely on synthetic data, utilizing a physics-based model for nematode motility, and we demonstrate the model's ability to generalize from simulations to experimental videos.

show abstract

Deep learning for robust and flexible tracking in behavioral studies for C. elegans

Cited by 22 publications

References 61 publications

Deep-Worm-Tracker: Deep Learning Methods for Accurate Detection and Tracking for Behavioral Studies in C. elegans

Deep-Worm-Tracker: Deep Learning Methods for Accurate Detection and Tracking for Behavioral Studies in C. elegans

A robotic system for automated genetic manipulation and analysis ofCaenorhabditis elegans

Fast spline detection in high density microscopy data

Contact Info

Product

Resources

About