DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci

Vičar, Tomáš; Gumulec, Jaromír; Kolář, Radim; Kopečná, Olga; Pagáčová, Eva; Falková, Iva; Falk, Martin

doi:10.1016/j.csbj.2021.11.019

Cited by 9 publications

(23 citation statements)

References 101 publications

(171 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several groups 32 , 59 , 60 successfully developed deep learning methods in automated foci scoring in different types of cells and yielded very good performances as mentioned in the introduction section. Moreover, Vicar et al 68 studied using 3D data and in heterogeneous cell populations. Previous studies 32 , 59 , 60 , 68 neither used YOLO algorithm for object detection nor studied in human blood samples.…”

Section: Discussionmentioning

confidence: 99%

“…Moreover, Vicar et al 68 studied using 3D data and in heterogeneous cell populations. Previous studies 32 , 59 , 60 , 68 neither used YOLO algorithm for object detection nor studied in human blood samples. FociRad, the two-stage learning method of the YOLO v4 for MNC and γ-H2AX foci detection, was applied in whole blood and yielded a very good F1 score of 0.911 (min 0.762, max 0.933).…”

Section: Discussionmentioning

confidence: 99%

“…In the same year, Vicar et al . 68 proposed a deep learning-based fully-automated method for 53BP1 and γ-H2AX foci counting called ‘DeepFoci’ in cell lines (NHDF, U-87) and primary cell cultures irradiated with 1–4 Gy gamma-rays. U-Net CNN 64 was used for nucleus segmentation and foci detection, followed by a Maximally Stable Extremal Region (MSER) algorithm 69 for segmentation of detected foci.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A deep learning model (FociRad) for automated detection of γ-H2AX foci and radiation dose estimation

Wanotayan

Chousangsuntorn

Petisiwaveth

et al. 2022

Sci Rep

View full text Add to dashboard Cite

DNA double-strand breaks (DSBs) are the most lethal form of damage to cells from irradiation. γ-H2AX (phosphorylated form of H2AX histone variant) has become one of the most reliable and sensitive biomarkers of DNA DSBs. However, the γ-H2AX foci assay still has limitations in the time consumed for manual scoring and possible variability between scorers. This study proposed a novel automated foci scoring method using a deep convolutional neural network based on a You-Only-Look-Once (YOLO) algorithm to quantify γ-H2AX foci in peripheral blood samples. FociRad, a two-stage deep learning approach, consisted of mononuclear cell (MNC) and γ-H2AX foci detections. Whole blood samples were irradiated with X-rays from a 6 MV linear accelerator at 1, 2, 4 or 6 Gy. Images were captured using confocal microscopy. Then, dose–response calibration curves were established and implemented with unseen dataset. The results of the FociRad model were comparable with manual scoring. MNC detection yielded 96.6% accuracy, 96.7% sensitivity and 96.5% specificity. γ-H2AX foci detection showed very good F1 scores (> 0.9). Implementation of calibration curve in the range of 0–4 Gy gave mean absolute difference of estimated doses less than 1 Gy compared to actual doses. In addition, the evaluation times of FociRad were very short (< 0.5 min per 100 images), while the time for manual scoring increased with the number of foci. In conclusion, FociRad was the first automated foci scoring method to use a YOLO algorithm with high detection performance and fast evaluation time, which opens the door for large-scale applications in radiation triage.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A deep learning model (FociRad) for automated detection of γ-H2AX foci and radiation dose estimation

Wanotayan

Chousangsuntorn

Petisiwaveth

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Many architectures have been proposed for various tasks, with CNNs being commonly used. DeepFoci 9 utilizes U-Net, a specific CNN architecture to detect and segment foci and nuclei with promising accuracy in a fully unsupervised manner. A combined approach in 10 uses classical techniques to detect regions of interest, extracts an over abundant set of features using filters and feeds these results into a classifier.…”

Section: Introductionmentioning

confidence: 99%

“…First, excluding manual classification (e.g., in Fiji) or generic approaches (e.g., CellProfiler), the identification of SGs, or other cytoplasmic foci with significant cytoplasmic background signal has seen very few tailored automated approaches. Second, as reported in an exhaustive study on bioimaging informatics tools 14 , overall usability is a significant hurdle for the use of automated methods, making CellProfiler one of most popular, yet not necessarily optimal 9 methods. Finally, in biological publications the issue of human variability and error is often not addressed.…”

Section: Introductionmentioning

confidence: 99%

HARLEY mitigates user bias and facilitates efficient quantification and co-localization analyses of foci in yeast fluorescence images

Shabanov

Buchan

2022

Sci Rep

View full text Add to dashboard Cite

Quantification of cellular structures in fluorescence microscopy data is a key means of understanding cellular function. Unfortunately, numerous cellular structures present unique challenges in their ability to be unbiasedly and accurately detected and quantified. In our studies on stress granules in yeast, users displayed a striking variation of up to 3.7-fold in foci calls and were only able to replicate their results with 62–78% accuracy, when re-quantifying the same images. To facilitate consistent results we developed HARLEY (Human Augmented Recognition of LLPS Ensembles in Yeast), a customizable software for detection and quantification of stress granules in S. cerevisiae. After a brief model training on ~ 20 cells the detection and quantification of foci is fully automated and based on closed loops in intensity contours, constrained only by the a priori known size of the features of interest. Since no shape is implied, this method is not limited to round features, as is often the case with other algorithms. Candidate features are annotated with a set of geometrical and intensity-based properties to train a kernel Support Vector Machine to recognize features of interest. The trained classifier is then used to create consistent results across datasets. For less ambiguous foci datasets, a parametric selection is available. HARLEY is an intuitive tool aimed at yeast microscopy users without much technical expertise. It allows batch processing of foci detection and quantification, and the ability to run various geometry-based and pixel-based colocalization analyses to uncover trends or correlations in foci-related data. HARLEY is open source and can be downloaded from https://github.com/lnilya/harley.

show abstract

Advanced image-free analysis of the nano-organization of chromatin and other biomolecules by Single Molecule Localization Microscopy (SMLM)

Weidner¹,

Neitzel²,

Gote³

et al. 2023

Computational and Structural Biotechnology Journal

Self Cite

View full text Add to dashboard Cite

DeepFoci: Deep learning-based algorithm for fast automatic analysis of DNA double-strand break ionizing radiation-induced foci

Abstract: Graphical abstract

Cited by 9 publications

References 101 publications

A deep learning model (FociRad) for automated detection of γ-H2AX foci and radiation dose estimation

A deep learning model (FociRad) for automated detection of γ-H2AX foci and radiation dose estimation

HARLEY mitigates user bias and facilitates efficient quantification and co-localization analyses of foci in yeast fluorescence images

Advanced image-free analysis of the nano-organization of chromatin and other biomolecules by Single Molecule Localization Microscopy (SMLM)

Contact Info

Product

Resources

About