Nucleus segmentation across imaging experiments: The 2018 Data Science Bowl
By Juan C. Caicedo, Allen Goodman, Kyle W. Karhohs, B. Cimini, J. Ackerman, Marzieh Haghighi, Cherkeng Heng, Tim Becker, M. Doan, Claire McQuin and 3 more. 2019
781 citations recorded by the source corpus.
Segmenting the nuclei of cells in microscopy images is often the first step in the quantitative analysis of imaging data for biological and biomedical applications. Many bioimage analysis tools can segment nuclei in images but need to be selected and configured for every experiment. The 2018 Data Science Bowl attracted 3,891 teams worldwide to make the first attempt to build a segmentation method that could be applied to any two-dimensional light microscopy image of stained nuclei across experiments, with no human interaction. Top participants in the challenge succeeded in this task, developing deep-learning-based models that identified cell nuclei across many image types and experimental conditions without the need to manually adjust segmentation parameters. This represents an important step toward configuration-free bioimage analysis software tools. The 2018 Data Science Bowl challenged competitors to develop an accurate tool for segmenting stained nuclei from diverse light microscopy images. The winners deployed innovative deep-learning strategies to realize configuration-free segmentation.
Subfields: Biophysics, Media Technology, Molecular Biology, Evaluation & Benchmarks, Medical Imaging
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Biophysics
- Marius Pachitariu
- C. Stringer
- Guangming Lu
- Alexander V. Buslaev
- Alexander Kirillov
- Robert Haase
- Bingzhi Chen
- Deborah Schmidt
In Media Technology
- Alexandr A Kalinin
- Ross B. Girshick
- Tete Xiao
- Marius Pachitariu
- C. Stringer
- Alexander V. Buslaev
- Vladimir I. Iglovikov
- Alexander C. Berg
In Molecular Biology
- C. Stringer
- Marius Pachitariu
- Florian Jug
- Ross B. Girshick
- Piotr Doll'ar
- Alexander C. Berg
- Hugo Touvron
- Herv'e J'egou
In Evaluation & Benchmarks
- Ross B. Girshick
- Alexander Kirillov
- Piotr Doll'ar
- Hanzi Mao
- Tete Xiao
- Alexander C. Berg
- Colin P. Elkin
- Alexander V. Buslaev
In Medical Imaging
Citations in the corpus
- Segment Anything (2023, background): “ADE20K[117] BBBC038v1[12] Cityscapes[25] DOORS[80] DRAM[24] EgoHOS[113] GTEA[34,63] Hypersim[86] IBD[17] iShape[111] LVIS[44] NDD20[100] NDISPark[22,23] OVIS[81] PPDLS[74] Plittersdorf[46] STREETS[91] TimberSeg[38] TrashCan[52] VISOR[28,27] WoodScape[112] PIDRay[104] ZeroWaste-f[6] Figure 8: Samples”
- Albumentations: fast and flexible image augmentations (2018, methodology): “Multiple annotations can be used for improving segmentation performance, as was recently demonstrated in the biological image segmentation task by the winning team of the Kaggle 2018 Data Science Bowl [46].”
- U-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications (2020, methodology): “Example images from Kaggle data science bowl 2018 [273].”
- TransAttUnet: Multi-Level Attention-Guided U-Net With Transformer for Medical Image Segmentation (2021, methodology): “Moreover, the proposed TransAttUnet is also evaluated on the Clean-CC-CCII dataset [34], 2018 Data Science Bowl (Bowl) dataset [35], and the Gland Segmentation (GlaS) dataset [19].”
- SAM 2: Segment Anything in Images and Videos (2024, neutral): “…23 datasets: LVIS (Gupta et al., 2019), ADE20K (Zhou et al., 2019), Hypersim (Roberts et al., 2021), Cityscapes (Cordts et al., 2016), BBBC038v1 (Caicedo et al., 2019), DOORS (Pugliatti & Topputo, 2022), DRAM (Cohen et al., 2022), EgoHOS (Zhang et al., 2022), GTEA (Fathi et al., 2011; Li et…”
- DS-TransUNet: Dual Swin Transformer U-Net for Medical Image Segmentation (2021, methodology): “2018 Data Science Bowl: The dataset is from 2018 Data Science Bowl challenge [42] and used to find the nuclei in divergent images, including 670 images in total.”
- DCSAU-Net: A Deeper and More Compact Split-Attention U-Net for Medical Image Segmentation (2022, methodology): “3) Experimental analysis is conducted with four different medical imagesegmentationdatasets, including2018Data Science Bowl [14], ISIC-2018Lesion BoundarySegmen-tation [ 15, 16], CVC-ClinicDB [17], and a multi-class segmentation dataset: SegPC-2021 [18].”
- Metrics Reloaded: Recommendations for image analysis validation (2022, neutral): “cell instance segmentation [17]), recent clinical initiatives advocate for cutoff-agnostic validation arguing that cutoffs are often over-optimized on a specific data set and that associated results are not transferable across study cohorts (e.”
- Towards a guideline for evaluation metrics in medical image segmentation (2022, neutral): “[35], endoscopy–endoscopic colonoscopy frames for polyp detection [36], fundus photography–vessel extraction in retinal images [37], dermoscopy–skin lesion segmentation for melanoma detection in dermoscopy images [38].”
- EMCAD: Efficient Multi-Scale Convolutional Attention Decoding for Medical Image Segmentation (2024, methodology): “We resize images to 352 × 352 and use a multi-scale { 0.75, 1.0, 1.25 } training strategy with a gradient clip limit of 0.5 for ClinicDB [3], Kvasir [29], ColonDB [51], ETIS [51], BKAI [40], ISIC17 [15], and ISIC18 [15], while we resize images to 256 × 256 for BUSI [1], EM [6], and DSB18 [4].”
- TrackMate 7: integrating state-of-the-art segmentation algorithms into tracking pipelines (2022, neutral): “16 ).”
- Cellpose 2.0: how to train your own model (2022, methodology): “Briefly this dataset consisted of images from BBBC038v1 [52, 53], BBBC039v1 [52], MoNuSeg [54], and ISBI 2009 [55].”