ESQmodel: Biologically Informed Evaluation of 2-D Cell Segmentation Quality in Multiplexed Tissue Images

Overview

Journal Bioinformatics

Publisher Oxford University Press

Specialty Biology

Date 2023 Dec 28

PMID 38152895

Authors

Eric Lee

Dongkyu Lee

Wayne Fan

Andrew Lytle

YuXiang Fu

David W Scott

Christian Steidl

Samuel Aparicio

Andrew Roth

Affiliations

Soon will be listed here.

Abstract

Motivation: Single cell segmentation is critical in the processing of spatial omics data to accurately perform cell type identification and analyze spatial expression patterns. Segmentation methods often rely on semi-supervised annotation or labeled training data which are highly dependent on user expertise. To ensure the quality of segmentation, current evaluation strategies quantify accuracy by assessing cellular masks or through iterative inspection by pathologists. While these strategies each address either the statistical or biological aspects of segmentation, there lacks a unified approach to evaluating segmentation accuracy.

Results: In this article, we present ESQmodel, a Bayesian probabilistic method to evaluate single cell segmentation using expression data. By using the extracted cellular data from segmentation and a prior belief of cellular composition as input, ESQmodel computes per cell entropy to assess segmentation quality by how consistent cellular expression profiles match with cell type expectations.

Availability And Implementation: Source code is available on Github at: https://github.com/Roth-Lab/ESQmodel.

References

Jackson H, Fischer J, Zanotelli V, Ali H, Mechera R, Soysal S . The single-cell pathology landscape of breast cancer. Nature. 2020; 578(7796):615-620. DOI: 10.1038/s41586-019-1876-x. View

Ali H, Jackson H, Zanotelli V, Danenberg E, Fischer J, Bardwell H . Imaging mass cytometry and multiplatform genomics define the phenogenomic landscape of breast cancer. Nat Cancer. 2022; 1(2):163-175. DOI: 10.1038/s43018-020-0026-6. View

Giesen C, Wang H, Schapiro D, Zivanovic N, Jacobs A, Hattendorf B . Highly multiplexed imaging of tumor tissues with subcellular resolution by mass cytometry. Nat Methods. 2014; 11(4):417-22. DOI: 10.1038/nmeth.2869. View

Goltsev Y, Samusik N, Kennedy-Darling J, Bhate S, Hale M, Vazquez G . Deep Profiling of Mouse Splenic Architecture with CODEX Multiplexed Imaging. Cell. 2018; 174(4):968-981.e15. PMC: 6086938. DOI: 10.1016/j.cell.2018.07.010. View

Lee I, Hyung Kim S, Song H, Park S . The molecular basis for the generation of Hodgkin and Reed-Sternberg cells in Hodgkin's lymphoma. Int J Hematol. 2003; 77(4):330-5. DOI: 10.1007/BF02982639. View

Bajcsy P, Cardone A, Chalfoun J, Halter M, Juba D, Kociolek M . Survey statistics of automated segmentations applied to optical imaging of mammalian cells. BMC Bioinformatics. 2015; 16:330. PMC: 4608288. DOI: 10.1186/s12859-015-0762-2. View

Bankhead P, Loughrey M, Fernandez J, Dombrowski Y, McArt D, Dunne P . QuPath: Open source software for digital pathology image analysis. Sci Rep. 2017; 7(1):16878. PMC: 5715110. DOI: 10.1038/s41598-017-17204-5. View

Greenwald N, Miller G, Moen E, Kong A, Kagel A, Dougherty T . Whole-cell segmentation of tissue images with human-level performance using large-scale data annotation and deep learning. Nat Biotechnol. 2021; 40(4):555-565. PMC: 9010346. DOI: 10.1038/s41587-021-01094-0. View

Levine J, Simonds E, Bendall S, Davis K, Amir E, Tadmor M . Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis. Cell. 2015; 162(1):184-97. PMC: 4508757. DOI: 10.1016/j.cell.2015.05.047. View

10.

Stringer C, Wang T, Michaelos M, Pachitariu M . Cellpose: a generalist algorithm for cellular segmentation. Nat Methods. 2020; 18(1):100-106. DOI: 10.1038/s41592-020-01018-x. View

11.

Lee E, Chern K, Nissen M, Wang X, Huang C, Gandhi A . SpatialSort: a Bayesian model for clustering and cell population annotation of spatial proteomics data. Bioinformatics. 2023; 39(39 Suppl 1):i131-i139. PMC: 10311307. DOI: 10.1093/bioinformatics/btad242. View

12.

Schneider C, Rasband W, Eliceiri K . NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012; 9(7):671-5. PMC: 5554542. DOI: 10.1038/nmeth.2089. View

13.

Hoch T, Schulz D, Eling N, Gomez J, Levesque M, Bodenmiller B . Multiplexed imaging mass cytometry of the chemokine milieus in melanoma characterizes features of the response to immunotherapy. Sci Immunol. 2022; 7(70):eabk1692. DOI: 10.1126/sciimmunol.abk1692. View

14.

Berg S, Kutra D, Kroeger T, Straehle C, Kausler B, Haubold C . ilastik: interactive machine learning for (bio)image analysis. Nat Methods. 2019; 16(12):1226-1232. DOI: 10.1038/s41592-019-0582-9. View

15.

Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T . Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012; 9(7):676-82. PMC: 3855844. DOI: 10.1038/nmeth.2019. View

16.

Angelo M, Bendall S, Finck R, Hale M, Hitzman C, Borowsky A . Multiplexed ion beam imaging of human breast tumors. Nat Med. 2014; 20(4):436-42. PMC: 4110905. DOI: 10.1038/nm.3488. View

17.

Stirling D, Swain-Bowden M, Lucas A, Carpenter A, Cimini B, Goodman A . CellProfiler 4: improvements in speed, utility and usability. BMC Bioinformatics. 2021; 22(1):433. PMC: 8431850. DOI: 10.1186/s12859-021-04344-9. View

18.

Caicedo J, Cooper S, Heigwer F, Warchal S, Qiu P, Molnar C . Data-analysis strategies for image-based cell profiling. Nat Methods. 2017; 14(9):849-863. PMC: 6871000. DOI: 10.1038/nmeth.4397. View

19.

Chen H, Murphy R . Evaluation of cell segmentation methods without reference segmentations. Mol Biol Cell. 2022; 34(6):ar50. PMC: 10208095. DOI: 10.1091/mbc.E22-08-0364. View

20.

Gerdes M, Sevinsky C, Sood A, Adak S, Bello M, Bordwell A . Highly multiplexed single-cell analysis of formalin-fixed, paraffin-embedded cancer tissue. Proc Natl Acad Sci U S A. 2013; 110(29):11982-7. PMC: 3718135. DOI: 10.1073/pnas.1300136110. View