Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Artificial Intelligence for CELLSEARCH Image Analysis.

Methods in molecular biology (Clifton, N.J.)·2026
Same author

Flow Enabled Target Capture Halbach-based magnetic enrichment increases circulating tumor cell capture from blood in metastatic cancer patients.

Molecular oncology·2026
Same author

Phenotypic and genotypic characterization of single circulating tumor cells in the follow-up of high-grade serous ovarian cancer.

Molecular oncology·2025
Same author

Longitudinal Analysis of Manually and Automatically Classified Circulating Tumor Biomarkers and their Prediction of Survival in Metastatic Colorectal Cancer.

Clinical pharmacology and therapeutics·2025
Same author

PSA Secretion from Single Circulating Tumor Cells of Metastatic Castration-Naïve Prostate Cancer Patients.

Cancer research communications·2025
Same author

Performance comparison of streptavidin magnetic beads for epcam expressing cancer cell lines for circulating tumor cell (CTC) enrichment in a flow-through immunomagnetic system.

PloS one·2025

Related Experiment Video

Updated: Sep 6, 2025

Visualization, Quantification, and Mapping of Immune Cell Populations in the Tumor Microenvironment
11:00

Visualization, Quantification, and Mapping of Immune Cell Populations in the Tumor Microenvironment

Published on: March 25, 2020

17.2K

StarDist Image Segmentation Improves Circulating Tumor Cell Detection.

Michiel Stevens1, Afroditi Nanou1, Leon W M M Terstappen1

  • 1Medical Cell Biophysics Group, Techmed Center, Faculty of Science and Technology, University of Twente, 7500 AE Enschede, The Netherlands.

Cancers
|June 24, 2022
PubMed
Summary

StarDist, a deep learning method, significantly improves circulating tumor cell (CTC) detection compared to CellSearch, especially in dense samples. This advanced segmentation ensures more accurate CTC enumeration for cancer diagnostics.

Keywords:
ACCEPTCellSearchStarDistcirculating tumor cell (CTC)diagnostic leukapheresis (DLA)image segmentation

More Related Videos

Semi-automatic PD-L1 Characterization and Enumeration of Circulating Tumor Cells from Non-small Cell Lung Cancer Patients by Immunofluorescence
10:29

Semi-automatic PD-L1 Characterization and Enumeration of Circulating Tumor Cells from Non-small Cell Lung Cancer Patients by Immunofluorescence

Published on: August 14, 2019

10.7K
Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
09:31

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

11.7K

Related Experiment Videos

Last Updated: Sep 6, 2025

Visualization, Quantification, and Mapping of Immune Cell Populations in the Tumor Microenvironment
11:00

Visualization, Quantification, and Mapping of Immune Cell Populations in the Tumor Microenvironment

Published on: March 25, 2020

17.2K
Semi-automatic PD-L1 Characterization and Enumeration of Circulating Tumor Cells from Non-small Cell Lung Cancer Patients by Immunofluorescence
10:29

Semi-automatic PD-L1 Characterization and Enumeration of Circulating Tumor Cells from Non-small Cell Lung Cancer Patients by Immunofluorescence

Published on: August 14, 2019

10.7K
Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
09:31

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

11.7K

Area of Science:

  • Biomedical Engineering
  • Computational Biology
  • Cancer Research

Background:

  • Circulating tumor cell (CTC) enumeration is crucial for cancer diagnosis and monitoring.
  • Current methods like CellSearch face challenges with high cell density samples.
  • Accurate segmentation of CTCs is essential for reliable enumeration.

Purpose of the Study:

  • To evaluate StarDist, a deep learning segmentation method, as an alternative to CellSearch for CTC detection.
  • To compare the performance of StarDist and CellSearch in both whole blood and diagnostic leukapheresis (DLA) samples.
  • To assess the impact of cell density on segmentation accuracy for both methods.

Main Methods:

  • Retrospective analysis of CellSearch image archives from 533 whole blood and 601 DLA samples.
  • Segmentation of images using both CellSearch algorithm and StarDist deep learning method.
  • Visual inspection and quantitative comparison of segmentation results by operators.

Main Results:

  • StarDist segmented 99.95% of CTCs identified by CellSearch in blood samples, with good outlines for 98.3%.
  • StarDist identified 10% more CTCs than CellSearch in blood samples and 20% more in DLA samples.
  • CellSearch segmentation failed in high-density DLA samples, producing excessively large segmentations, while StarDist maintained performance.

Conclusions:

  • StarDist offers superior performance for CTC segmentation compared to CellSearch, particularly in high cell density samples.
  • Deep learning-based segmentation with StarDist enhances CTC enumeration accuracy.
  • StarDist is a promising tool for improving CTC analysis in complex clinical samples.