On Improving an Already Competitive Segmentation Algorithm for the Cell Tracking Challenge - Lessons Learned

Scherr, Tim; Loeffler, K; Neumann, Oliver; Mikut, Ralf

doi:10.1101/2021.06.26.450019

Cited by 8 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main difference between the models is that U-Net3-BF-1 (figure 7 h ) produces cells with more consistent and smooth shapes, similar to the ground truth (figure 7 b ) while RF-BF-2 (figure 7 g ) creates rugged edges and also detects many small fragmented objects far from the cells. Numerically, the quality of bright-field detection (table 1, F 1 score = 0.89) is somewhat lower than the current state of the art solutions in the cell tracking challenge [65] when compared to the most similar dataset of DIC-HeLa cells ( F 1 score = 0.93) [65–68]. However, it must be noted that such small differences could be easily caused by differences in imaging modes, magnifications, cell line morphology or amount of training data among other parameters.…”

Section: Resultsmentioning

confidence: 99%

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M ₄ muscarinic receptors?

et al. 2022

View full text Add to dashboard Cite

M 4 muscarinic acetylcholine receptor is a G protein-coupled receptor (GPCR) that has been associated with alcohol and cocaine abuse, Alzheimer's disease, and schizophrenia which makes it an interesting drug target. For many GPCRs, the high-affinity fluorescence ligands have expanded the options for high-throughput screening of drug candidates and serve as useful tools in fundamental receptor research. Here, we explored two TAMRA-labelled fluorescence ligands, UR-MK342 and UR-CG072, for development of assays for studying ligand-binding properties to M 4 receptor. Using budded baculovirus particles as M 4 receptor preparation and fluorescence anisotropy method, we measured the affinities and binding kinetics of both fluorescence ligands. Using the fluorescence ligands as reporter probes, the binding affinities of unlabelled ligands could be determined. Based on these results, we took a step towards a more natural system and developed a method using live CHO-K1-hM 4 R cells and automated fluorescence microscopy suitable for the routine determination of unlabelled ligand affinities. For quantitative image analysis, we developed random forest and deep learning-based pipelines for cell segmentation. The pipelines were integrated into the user-friendly open-source Aparecium software. Both image analysis methods were suitable for measuring fluorescence ligand saturation binding and kinetics as well as for screening binding affinities of unlabelled ligands.

show abstract

Section: Resultsmentioning

confidence: 99%

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M ₄ muscarinic receptors?

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Fluo-N2DL-HeLa. The public Fluo-N2DL-HeLa dataset [53] [46] 96 Our H-EMD TUG-AT [10] CALT-US [13] BGU-IL [19] KIT-Sch-GE [18] DKFZ-GE [48] MU-Ba-US [49] UNSW-AU [50] UVA-NL [15] FR-Ro-GE [6] RWTH-GE [11] BRF-GE [51] KTH-SE [46] Threshold values Scores (%) Fungus. The in-house Fungus dataset [52], [55], [56] contains 4 3D electron microscopy images for segmenting fungus cells captured from body tissues of ants, whose 2D slices are of 853 × 877 pixels each.…”

Section: Methodsmentioning

confidence: 99%

“…Another line of work aimed to incorporate instance-level topological properties by transforming them into training labels and corresponding object functions [16], [34]. In distance map methods [16], [18], [35], the Euclidean distances between each instance pixel to the nearest instance boundary are utilized as target labels. Graham et al [17] proposed to use both horizontal and vertical distances between each instance pixel to the corresponding instance mass centers.…”

Section: Related Workmentioning

confidence: 99%

“…The strategies of the second type aim to utilize instance morphological properties to distinguish different instances [11]- [13]. Some of the best known methods are watershed-based [14], directly applying watershed as a post-processing [11], [15] or watershedinspired deep learning methods such as distance maps [16]- [18] and gradient maps [12], which show great performances on a wide range of biomedical image datasets.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

H-EMD: A Hierarchical Earth Mover’s Distance Method for Instance Segmentation

Liang¹,

Zhang²,

Ding³

et al. 2022

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Deep learning (DL) based semantic segmentation methods have achieved excellent performance in biomedical image segmentation, producing high quality probability maps to allow extraction of rich instance information to facilitate good instance segmentation. While numerous efforts were put into developing new DL semantic segmentation models, less attention was paid to a key issue of how to effectively explore their probability maps to attain the best possible instance segmentation. We observe that probability maps by DL semantic segmentation models can be used to generate many possible instance candidates, and accurate instance segmentation can be achieved by selecting from them a set of "optimized" candidates as output instances. Further, the generated instance candidates form a well-behaved hierarchical structure (a forest), which allows selecting instances in an optimized manner. Hence, we propose a novel framework, called hierarchical earth mover's distance (H-EMD), for instance segmentation in biomedical 2D+time videos and 3D images, which judiciously incorporates consistent instance selection with semantic-segmentation-generated probability maps. H-EMD contains two main stages. (1) Instance candidate generation: capturing instance-structured information in probability maps by generating many instance candidates in a forest structure. (2) Instance candidate selection: selecting instances from the candidate set for final instance segmentation. We formulate a key instance selection problem on the instance candidate forest as an optimization problem based on the earth mover's distance (EMD), and solve it by integer linear programming. Extensive experiments on eight biomedical video or 3D datasets demonstrate that H-EMD consistently boosts DL semantic segmentation models and is highly competitive with stateof-the-art methods.

show abstract

“…Usually, the default settings are sufficient but the batch size may be reduced if memory availability is limited. A detailed description of the training process is provided by Scherr et al 22,23…”

Section: Trainingmentioning

confidence: 99%

microbeSEG: A deep learning software tool with OMERO data management for efficient and accurate cell segmentation

Scherr

Seiffarth

Wollenhaupt

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In biotechnology, cell growth is one of the most important properties for the characterization and optimization of microbial cultures. Novel live-cell imaging methods are leading to an ever better understanding of cell cultures and their development. The key for analyzing acquired data is accurate and automated cell segmentation at the single-cell level. Therefore, we present microbeSEG, a user-friendly cell segmentation tool with OMERO data management. microbeSEG utilizes a state-of-the-art deep learning-based segmentation method and can be used for instance segmentation of a wide range of cell morphologies and imaging techniques, e.g., phase contrast or fluorescence microscopy. The main focus of microbeSEG is a comprehensible, easy, efficient, and complete workflow from creation of training data to final application of the trained segmentation model. We demonstrate that accurate cell segmentation results can be obtained within 45 minutes of user time. Utilizing public segmentation data sets or pre-labeling further accelerates the microbeSEG workflow. This opens the door for accurate and efficient data analysis of microbial cultures.

show abstract

On Improving an Already Competitive Segmentation Algorithm for the Cell Tracking Challenge - Lessons Learned

Cited by 8 publications

References 23 publications

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M ₄ muscarinic receptors?

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M ₄ muscarinic receptors?

H-EMD: A Hierarchical Earth Mover’s Distance Method for Instance Segmentation

microbeSEG: A deep learning software tool with OMERO data management for efficient and accurate cell segmentation

Contact Info

Product

Resources

About

On Improving an Already Competitive Segmentation Algorithm for the Cell Tracking Challenge - Lessons Learned

Cited by 8 publications

References 23 publications

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M 4 muscarinic receptors?

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M 4 muscarinic receptors?

H-EMD: A Hierarchical Earth Mover’s Distance Method for Instance Segmentation

microbeSEG: A deep learning software tool with OMERO data management for efficient and accurate cell segmentation

Contact Info

Product

Resources

About

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M ₄ muscarinic receptors?

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses—which way to go with measuring ligand binding to M ₄ muscarinic receptors?