SyMBac: Synthetic Micrographs for Accurate Segmentation of Bacterial Cells using Deep Neural Networks

Hardo, Georgeos; Noka, Maximilian; Bakshi, Somenath

doi:10.1101/2021.07.21.453284

Cited by 3 publications

(3 citation statements)

References 44 publications

(96 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Project name: SyMBac • Project homepage: https:// github. com/ georg eosha rdo/ SyMBac [47] • Project documentation and datasets: https:// symbac. readt hedocs.…”

Section: Supplementary Informationmentioning

confidence: 99%

Synthetic Micrographs of Bacteria (SyMBac) allows accurate segmentation of bacterial cells using deep neural networks

2022

View full text Add to dashboard Cite

Background Deep-learning–based image segmentation models are required for accurate processing of high-throughput timelapse imaging data of bacterial cells. However, the performance of any such model strictly depends on the quality and quantity of training data, which is difficult to generate for bacterial cell images. Here, we present a novel method of bacterial image segmentation using machine learning models trained with Synthetic Micrographs of Bacteria (SyMBac). Results We have developed SyMBac, a tool that allows for rapid, automatic creation of arbitrary amounts of training data, combining detailed models of cell growth, physical interactions, and microscope optics to create synthetic images which closely resemble real micrographs, and is capable of training accurate image segmentation models. The major advantages of our approach are as follows: (1) synthetic training data can be generated virtually instantly and on demand; (2) these synthetic images are accompanied by perfect ground truth positions of cells, meaning no data curation is required; (3) different biological conditions, imaging platforms, and imaging modalities can be rapidly simulated, meaning any change in one’s experimental setup no longer requires the laborious process of manually generating new training data for each change. Deep-learning models trained with SyMBac data are capable of analysing data from various imaging platforms and are robust to drastic changes in cell size and morphology. Our benchmarking results demonstrate that models trained on SyMBac data generate more accurate cell identifications and precise cell masks than those trained on human-annotated data, because the model learns the true position of the cell irrespective of imaging artefacts. We illustrate the approach by analysing the growth and size regulation of bacterial cells during entry and exit from dormancy, which revealed novel insights about the physiological dynamics of cells under various growth conditions. Conclusions The SyMBac approach will help to adapt and improve the performance of deep-learning–based image segmentation models for accurate processing of high-throughput timelapse image data.

show abstract

“…• Project name: SyMBac • Project homepage: https:// github. com/ georg eosha rdo/ SyMBac [47] • Project documentation and datasets: https:// symbac. readt hedocs.…”

Section: Supplementary Informationmentioning

confidence: 99%

Synthetic Micrographs of Bacteria (SyMBac) allows accurate segmentation of bacterial cells using deep neural networks

2022

View full text Add to dashboard Cite

show abstract

“…Synthetic data have been used before, but usually as a way to increase the amount of training data, that is, available 27 or to ensure that all variance, that is present in the real world is also present in the training data 28 . In fact, one could argue that many forms of data augmentation 25 synthesize the data to some degree.…”

Section: Discussionmentioning

confidence: 99%

Impact of imperfection in medical imaging data on deep learning‐based segmentation performance: An experimental study using synthesized data

et al. 2023

View full text Add to dashboard Cite

BackgroundClinical data used to train deep learning models are often not clean data. They can contain imperfections in both the imaging data and the corresponding segmentations.PurposeThis study investigates the influence of data imperfections on the performance of deep learning models for parotid gland segmentation. This was done in a controlled manner by using synthesized data. The insights this study provides may be used to make deep learning models better and more reliable.MethodsThe data were synthesized by using the clinical segmentations, creating a pseudo ground‐truth in the process. Three kinds of imperfections were simulated: incorrect segmentations, low image contrast, and artifacts in the imaging data. The severity of each imperfection was varied in five levels. Models resulting from training sets from each of the five levels were cross‐evaluated with test sets from each of the five levels.ResultsUsing synthesized data led to almost perfect parotid gland segmentation when no error was added. Lowering the quality of the parotid gland segmentations used for training substantially lowered the model performance. Additionally, lowering the image quality of the training data by decreasing the contrast or introducing artifacts made the resulting models more robust to data containing those respective kinds of data imperfection.ConclusionThis study demonstrated the importance of good‐quality segmentations for deep learning training and it shows that using low‐quality imaging data for training can enhance the robustness of the resulting models.

show abstract

“…For backwards compatibility, the version of SyMBac used in this paper has been frozen and included in this repository. Sample datasets, including instrumental point spread functions, microscope images of membrane stained cells and microcolonies, synthetic benchmarking data, and mother machine data have been uploaded to https://zenodo.org/records/10525762 70 .…”

Section: Data and Code Availabilitymentioning

confidence: 99%

Quantitative microbiology with widefield microscopy: navigating optical artefacts for accurate interpretations

Hardo

Bakshi

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Live cell imaging of microbial cells with microscopy has revolutionised quantitative microbiology. Micrographs are one of the most information-rich data types captured about a microbe, allowing quantification of the size and morphology of individual cells and their gene-expression over time. However, an optical microscope is a diffraction-limited system, and the comparable size of the point spread function of the microscope to the size of a microbial cell can lead to imaging artefacts which corrupt and bias the data. Additionally, the comparable thickness of a microbe to the depth of field of the microscope means that the 2D image contains compressed, projected 3D information. This makes it difficult to extract the underlying 3D distribution of photon emitters. For unknown distributions, the problem can be as ill-posed as a deconvolution problem, usually not having a unique solution. Together, the diffraction and projection effects affect our ability to accurately quantify the size and shape of microbial cells from their images and their contents from intensity measurements. In this paper, we use a mixture of simulations and experiments of microscopic image formation of microbial cells to illustrate the effects of diffraction and projection on cell segmentation and signal quantification. We use targeted experiments to validate the predictions where possible. Finally, we use the knowledge of these effects to design experiments which can help to reduce the errors and biases in our analysis. Awareness of these effects and the approaches towards alleviating them will help to accurately quantify microbiology from microscopy data.

show abstract

SyMBac: Synthetic Micrographs for Accurate Segmentation of Bacterial Cells using Deep Neural Networks

Cited by 3 publications

References 44 publications

Synthetic Micrographs of Bacteria (SyMBac) allows accurate segmentation of bacterial cells using deep neural networks

Synthetic Micrographs of Bacteria (SyMBac) allows accurate segmentation of bacterial cells using deep neural networks

Impact of imperfection in medical imaging data on deep learning‐based segmentation performance: An experimental study using synthesized data

Quantitative microbiology with widefield microscopy: navigating optical artefacts for accurate interpretations

Contact Info

Product

Resources

About