StarDist Image Segmentation Improves Circulating Tumor Cell Detection

Stevens, Michiel; Nanou, Afroditi; Terstappen, Leon W.M.M.; Driemel, C; Stoecklein, Nikolas H.; Coumans, F.A.W.

doi:10.3390/cancers14122916

Cited by 31 publications

(24 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a large multicenter international study for BrCa CTC detection, a higher inter-reader variability was found in the early cancer setting, where a small number of CTCs was detected . Semiautomated CTC detection could be considerably improved by new image processing tools, such as deep-learning-based segmentation algorithms . Preanalytical variables (e.g., blood collection tubes, sample collection and handling, processing time, etc.)…”

Section: Challenges and Perspectives In Ctc Analyses For Medical Diag...mentioning

confidence: 99%

Circulating Tumor Cells in Cancer Diagnostics and Prognostics by Single-Molecule and Single-Cell Characterization

Chowdhury¹,

Cressiot

Parisi

et al. 2023

ACS Sens.

View full text Add to dashboard Cite

Circulating tumor cells (CTCs) represent an interesting source of biomarkers for diagnosis, prognosis, and the prediction of cancer recurrence, yet while they are extensively studied in oncobiology research, their diagnostic utility has not yet been demonstrated and validated. Their scarcity in human biological fluids impedes the identification of dangerous CTC subpopulations that may promote metastatic dissemination. In this Perspective, we discuss promising techniques that could be used for the identification of these metastatic cells. We first describe methods for isolating patient-derived CTCs and then the use of 3D biomimetic matrixes in their amplification and analysis, followed by methods for further CTC analyses at the single-cell and single-molecule levels. Finally, we discuss how the elucidation of mechanical and morphological properties using techniques such as atomic force microscopy and molecular biomarker identification using nanopore-based detection could be combined in the future to provide patients and their healthcare providers with a more accurate diagnosis.

show abstract

Section: Challenges and Perspectives In Ctc Analyses For Medical Diag...mentioning

confidence: 99%

Circulating Tumor Cells in Cancer Diagnostics and Prognostics by Single-Molecule and Single-Cell Characterization

Chowdhury¹,

Cressiot

Parisi

et al. 2023

ACS Sens.

View full text Add to dashboard Cite

show abstract

“…For each cell, mean protein intensity signal was measured and assigned into the cell protein expression level. After several signal preprocessing steps to normalise and remove outlier signal, a standard cell type clustering was applied using a common single-cell processing pipeline, scanpy (33,45).…”

Section: Phenoimager Ht Cell Type and Cell Community Analysismentioning

confidence: 99%

A robust Platform for Integrative Spatial Multi-omics Analysis to Map Immune Responses to SARS-CoV-2 infection in Lung Tissues

Tan

Grice

Trân

et al. 2023

Preprint

View full text Add to dashboard Cite

The SARS-CoV-2 (COVID-19) virus has caused a devastating global pandemic of respiratory illness. To understand viral pathogenesis, methods are available for studying dissociated cells in blood, nasal samples, bronchoalveolar lavage fluid, and similar, but a robust platform for deep tissue characterisation of molecular and cellular responses to virus infection in the lungs is still lacking. We developed an innovative spatial multi-omics platform to investigate COVID-19-infected lung tissues. Five tissue-profiling technologies were combined by a novel computational mapping methodology to comprehensively characterise and compare the transcriptome and targeted proteome of virus infected and uninfected tissues. By integrating spatial transcriptomics data (Visium, GeoMx and RNAScope) and proteomics data (CODEX and PhenoImager HT) at different cellular resolutions across lung tissues, we found strong evidence for macrophage infiltration and defined the broader microenvironment surrounding these cells. By comparing infected and uninfected samples, we found an increase in cytokine signalling and interferon responses at different sites in the lung and showed spatial heterogeneity in the expression level of these pathways. These data demonstrate that integrative spatial multi-omics platforms can be broadly applied to gain a deeper understanding of viral effects on cellular environments at the site of infection and to increase our understanding of the impact of SARS-CoV-2 on the lungs.

show abstract

“…As such, poor single cell segmentation can have dramatic and confounding effects on accurate cell type/state identification, akin to measuring doublets/aggregates of debris by conventional flow cytometry or scRNAseq. There are a number of published "end to end" pipelines for IMC data analysis (8)(9)(10)(11)(12) that utilise open source software for segmentation such as Ilastik (13) and CellProfiler (14,15), as well as StarDist (16) and IMC-specific approaches that utilise deep learning (17). There have also been attempts to use matched fluorescent images of the nuclei using DAPI co-staining to improve segmentation accuracy (18) as well as removing image noise (19,20).…”

Section: Introductionmentioning

confidence: 99%

OPTIMAL: An OPTimised Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration

Hunter

Nicorescu

Foster

et al. 2023

Preprint

View full text Add to dashboard Cite

Analysis of Imaging Mass Cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single cell segmentation and sub-optimal approaches for data visualisation and exploration. This can lead to inaccurate identification of cell phenotypes, states or spatial relationships compared to reference data from single cell suspension technologies. To this end we have developed the "OPTIMAL" framework to determine the best approaches for cell segmentation, parameter transformation, batch effect correction, data visualisation/clustering and spatial neighbourhood analysis. Using a panel of 27 metal-tagged antibodies recognising well characterised phenotypic and functional markers to stain the same FFPE human tonsil sample Tissue Microarray (TMA) over 12 temporally distinct batches we tested a total of four cell segmentation models, a range of different arcsinh cofactor parameter transformation values, five different dimensionality reduction algorithms and two clustering methods. Finally we assessed the optimal approach for performing neighbourhood analysis. We found that single cell segmentation was improved by the use of an Ilastik-derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using "classical" bi-variate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximised the statistical separation between negative and positive signal distributions and a simple Z-score normalisation step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out-performing Phenograph in terms of cell type identification. We also found that neighbourhood analysis was influenced by the method used for finding neighbouring cells with a "disc" pixel expansion outperforming a "bounding box" approach combined with the need for filtering objects based on size and image-edge location. Importantly OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates .FCS files from the segmentation output, allows for single cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists.

show abstract

StarDist Image Segmentation Improves Circulating Tumor Cell Detection

Cited by 31 publications

References 27 publications

Circulating Tumor Cells in Cancer Diagnostics and Prognostics by Single-Molecule and Single-Cell Characterization

Circulating Tumor Cells in Cancer Diagnostics and Prognostics by Single-Molecule and Single-Cell Characterization

A robust Platform for Integrative Spatial Multi-omics Analysis to Map Immune Responses to SARS-CoV-2 infection in Lung Tissues

OPTIMAL: An OPTimised Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration

Contact Info

Product

Resources

About