automated-analysis-of-c-elegans-fluorescence-images-using-segelegans

Automated Analysis of <em>C. elegans</em> Fluorescence Images using SegElegans

Microscopy, particularly of the fluorescent kind, is a frequently used tool in C. elegans research. The analysis of data from microscopy experiments can, ...

Universitat Politècnica de València

The automation of lifespan assays with  in standard Petri dishes is a challenging problem because there are several problems hindering detection such as occlusions at the plate edges, dirt accumulation, and worm aggregations. Moreover, determining whether a worm is alive or dead can be complex as they barely move during the last few days of their lives. This paper proposes a method combining traditional computer vision techniques with a live/dead  classifier based on convolutional and recurrent neural networks from low-resolution image sequences. In addition to proposing a new method to automate lifespan, the use of data augmentation techniques is proposed to train the network in the absence of large numbers of samples. The proposed method achieved small error rates (3.54% ± 1.30% per plate) with respect to the manual curve, demonstrating its feasibility.

Towards Lifespan Automation for  Based on Deep Learning: Analysing Convolutional and Recurrent Neural Networks for Dead or Live Classification.

Automatic tracking of  () in standard Petri dishes is challenging due to high-resolution image requirements when fully monitoring a Petri dish, but mainly due to potential losses of individual worm identity caused by aggregation of worms, overlaps and body contact. To date, trackers only automate tests for individual worm behaviors, canceling data when body contact occurs. However, essays automating contact behaviors still require solutions to this problem. In this work, we propose a solution to this difficulty using computer vision techniques. On the one hand, a skeletonization method is applied to extract skeletons in overlap and contact situations. On the other hand, new optimization methods are proposed to solve the identity problem during these situations. Experiments were performed with 70 tracks and 3779 poses (skeletons) of . Several cost functions with different criteria have been evaluated, and the best results gave an accuracy of 99.42% in overlapping with other worms and noise on the plate using the modified skeleton algorithm and 98.73% precision using the classical skeleton algorithm.

 Multi-Tracker Based on a Modified Skeleton Algorithm.

In recent decades, assays with the nematode  () have enabled great advances to be made in research on aging. However, performing these assays manually is a laborious task. To solve this problem, numerous  assay automation techniques are being developed to increase throughput and accuracy. In this paper, a method for predicting the lifespan of  nematodes using a bimodal neural network is proposed and analyzed. Specifically, the model uses the sequence of images and the count of live  up to the current day to predict the lifespan curve termination. This network has been trained using a simulator to avoid the labeling costs of training such a model. In addition, a method for estimating the uncertainty of the model predictions has been proposed. Using this uncertainty, a criterion has been analyzed to decide at what point the assay could be halted and the user could rely on the model's predictions. The method has been analyzed and validated using real experiments. The results show that uncertainty is reduced from the mean lifespan and that most of the predictions obtained do not present statistically significant differences with respect to the curves obtained manually.

Analysis of a  lifespan prediction method based on a bimodal neural network and uncertainty estimation.

Pose estimation of  in image sequences is challenging and even more difficult in low-resolution images. Problems range from occlusions, loss of worm identity, and overlaps to aggregations that are too complex or difficult to resolve, even for the human eye. Neural networks, on the other hand, have shown good results in both low-resolution and high-resolution images. However, training in a neural network model requires a very large and balanced dataset, which is sometimes impossible or too expensive to obtain. In this article, a novel method for predicting  poses in cases of multi-worm aggregation and aggregation with noise is proposed. To solve this problem we use an improved U-Net model capable of obtaining images of the next aggregated worm posture. This neural network model was trained/validated using a custom-generated dataset with a synthetic image simulator. Subsequently, tested with a dataset of real images. The results obtained were greater than 75% in precision and 0.65 with Intersection over Union (IoU) values.

Automatic segmentation of  skeletons in worm aggregations using improved U-Net in low-resolution image sequences.

Performing lifespan assays with  () nematodes manually is a time consuming and laborious task. Therefore, automation is necessary to increase productivity. In this paper, we propose a method to automate the counting of live  using deep learning. The survival curves of the experiment are obtained using a sequence formed by an image taken on each day of the assay. Solving this problem would require a very large labeled dataset; thus, to facilitate its generation, we propose a simplified image-based strategy. This simplification consists of transforming the real images of the nematodes in the Petri dish to a synthetic image, in which circular blobs are drawn on a constant background to mark the position of the . To apply this simplification method, it is divided into two steps. First, a Faster R-CNN network detects the , allowing its transformation into a synthetic image. Second, using the simplified image sequence as input, a regression neural network is in charge of predicting the count of live nematodes on each day of the experiment. In this way, the counting network was trained using a simple simulator, avoiding labeling a very large real dataset or developing a realistic simulator. Results showed that the differences between the curves obtained by the proposed method and the manual curves are not statistically significant for either short-lived N2 (p-value log rank test 0.45) or long-lived  (p-value log rank test 0.83) strains.

Automation of  lifespan assay using a simplified domain synthetic image-based neural network training strategy.

[This corrects the article DOI: 10.1016/j.csbj.2023.09.039.].

Corrigendum to "Towards generalization for  detection" [Comput Struct Biotechnol J 21 (2023) 4914-4922].

Caenorhabditis elegans is a great model for exploring organismal, cellular, and subcellular biology through optical and fluorescence microscopy, with its research applications steadily expanding. However, manual processing of numerous microscopic images is prone to errors and demands significant labor due to worms tendency to touch or cluster with each other. Here, we present a new system for segmenting whole-body instances of Caenorhabditis elegans in microscopic images (referred to as SegElegans), employing a combination of neural network architecture and conventional image processing techniques. Our method effectively overcomes previous challenges and resolves many instances of contact and overlap between worms in highly populated images in a timely manner. The results obtained show an average Intersection over Union value of 96.3% per worm and an average improvement of 6% over other existing methods for automated analysis of worm images. SegElegns is a user-friendly application for Caenorhabditis elegans segmentation that will benefit whole-worm phenotypic screenings essential for studying development, behavior, aging, and disease.

SegElegans: Instance segmentation using dual convolutional recurrent neural network decoder in Caenorhabditis elegans microscopic images.

Caenorhabditis elegans is a widely used animal model for researching new disease treatments. In recent years, automated methods have been developed to extract mobility phenotypes and analyse, using statistical methods, whether there are differences between control strains and disease model strains. However, these methods present certain limitations in detecting subtle and non-linear patterns. In this study, we propose a high-throughput screening method based on machine learning, using classifiers that provide a recovery percentage as a measure of treatment effect. We evaluate two main approaches: traditional machine learning models based on behavioral features extracted from the worm's skeleton using Tierpsy Tracker, and deep neural networks that directly analyse video sequences. The results indicate that a Random Forest classifier trained with features extracted by Tierpsy Tracker offers higher accuracy and explainability, making it more suitable than deep learning models for drug testing experiments. Finally, to assess the applicability of our method, we processed data from a published drug repurposing study on unc-80 mutants based on statistical methods. The results highlight the potential of machine learning models to enhance automated phenotypic screening in animal models, providing a more robust and quantitative evaluation of treatment effects by considering more complex and subtle patterns.

Article	Year
Towards Lifespan Automation for Based on Deep Learning: Analysing Convolutional and Recurrent Neural Networks for Dead or Live Classification. Sensors (Basel, Switzerland)\| PubMed ID: 34300683	2021
Multi-Tracker Based on a Modified Skeleton Algorithm. Sensors (Basel, Switzerland)\| PubMed ID: 34451062	2021
Analysis of a lifespan prediction method based on a bimodal neural network and uncertainty estimation. Computational and structural biotechnology journal\| PubMed ID: 36659931	2023
Automatic segmentation of skeletons in worm aggregations using improved U-Net in low-resolution image sequences. Heliyon\| PubMed ID: 37025880	2023
Automation of lifespan assay using a simplified domain synthetic image-based neural network training strategy. Computational and structural biotechnology journal\| PubMed ID: 37867965	2023
Corrigendum to "Towards generalization for detection" [Comput Struct Biotechnol J 21 (2023) 4914-4922]. Computational and structural biotechnology journal\| PubMed ID: 38948166	2024
SegElegans: Instance segmentation using dual convolutional recurrent neural network decoder in Caenorhabditis elegans microscopic images. Computers in biology and medicine\| PubMed ID: 40120179	2025
High-throughput behavioral screening in Caenorhabditis elegans using machine learning for drug repurposing. Scientific reports\| PubMed ID: 40681561	2025

Antonio Garcia Garvi

Biography

JoVE Journal Publications

Other Publications