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

Integrative plasma-to-spatial proteomics reveals fibroblast-associated signatures in liver metastatic breast cancer.

Cancer cell international·2026
Same author

Metered cryospray in patients with chronic bronchitis: a mechanistic randomised controlled study.

Thorax·2026
Same author

Promise and pragmatism of AI in global-scale digital pathology: pan-cancer approaches for clinical practice.

Journal of clinical pathology·2026
Same author

Improving Breast Cancer Outcomes by Enhanced Activities in Early Detection and Diagnosis: An Umbrella Review and Meta-Analyses of Randomised Controlled Trials in High-Income Contexts With Universal Healthcare Coverage.

Cancer control : journal of the Moffitt Cancer Center·2026
Same author

Type 3 innate lymphoid cells dominate the ILC compartment in endstage lung disease.

Frontiers in immunology·2026
Same author

Airway microbiome diversity, intra-mucosal bacteria, and spatial immunity in asthmatics and controls.

American journal of respiratory and critical care medicine·2026

Related Experiment Video

Updated: Jul 17, 2025

Generation and Expansion of Primary, Malignant Pleural Mesothelioma Tumor Lines
08:01

Generation and Expansion of Primary, Malignant Pleural Mesothelioma Tumor Lines

Published on: April 21, 2022

1.9K

Malignant Mesothelioma subtyping via sampling driven multiple instance prediction on tissue image and cell morphology

Mark Eastwood1, Silviu Tudor Marc2, Xiaohong Gao2

  • 1Tissue Image Analytics Center, University of Warwick, United Kingdom.

Artificial Intelligence in Medicine
|September 6, 2023
PubMed
Summary

This study introduces a novel multiple instance learning (MIL) approach for accurately subtyping malignant mesothelioma, a lethal cancer linked to asbestos exposure. The method offers continuous characterization, improving diagnostic objectivity and potentially patient outcomes.

Keywords:
Cancer subtypingComputational pathologyDeep learningMalignant MesotheliomaMultiple instance learning

More Related Videos

Orthotopic Implantation and Peripheral Immune Cell Monitoring in the II-45 Syngeneic Rat Mesothelioma Model
09:31

Orthotopic Implantation and Peripheral Immune Cell Monitoring in the II-45 Syngeneic Rat Mesothelioma Model

Published on: October 2, 2015

9.2K
DNA-barcode-based Multiplex Immunofluorescence Imaging to Analyze FFPE Specimens from Genetically Reprogrammed Murine Melanoma
09:52

DNA-barcode-based Multiplex Immunofluorescence Imaging to Analyze FFPE Specimens from Genetically Reprogrammed Murine Melanoma

Published on: June 6, 2025

267

Related Experiment Videos

Last Updated: Jul 17, 2025

Generation and Expansion of Primary, Malignant Pleural Mesothelioma Tumor Lines
08:01

Generation and Expansion of Primary, Malignant Pleural Mesothelioma Tumor Lines

Published on: April 21, 2022

1.9K
Orthotopic Implantation and Peripheral Immune Cell Monitoring in the II-45 Syngeneic Rat Mesothelioma Model
09:31

Orthotopic Implantation and Peripheral Immune Cell Monitoring in the II-45 Syngeneic Rat Mesothelioma Model

Published on: October 2, 2015

9.2K
DNA-barcode-based Multiplex Immunofluorescence Imaging to Analyze FFPE Specimens from Genetically Reprogrammed Murine Melanoma
09:52

DNA-barcode-based Multiplex Immunofluorescence Imaging to Analyze FFPE Specimens from Genetically Reprogrammed Murine Melanoma

Published on: June 6, 2025

267

Area of Science:

  • Oncology
  • Computational Pathology
  • Digital Health

Background:

  • Malignant mesothelioma is a rare, aggressive cancer often linked to asbestos exposure.
  • Accurate subtyping (Epithelioid, Sarcomatoid, Biphasic) is crucial for treatment but faces high inter-observer variability in histological assessment.
  • Transitional histological features complicate definitive subtyping, impacting patient management.

Purpose of the Study:

  • To develop an automated, objective method for subtyping malignant mesothelioma using deep learning.
  • To overcome the limitations of subjective histological subtyping and high inter-observer variability.
  • To enable continuous characterization of tumor subtypes and study heterogeneity.

Main Methods:

  • An end-to-end multiple instance learning (MIL) framework was developed for malignant mesothelioma subtyping.
  • An adaptive instance-based sampling scheme trained deep convolutional neural networks on image patches.
  • Instance representations were augmented with aggregate cellular morphology features from cell segmentation.

Main Results:

  • The proposed MIL approach accurately identified malignant mesothelial subtypes within tissue regions.
  • A continuous characterization of sarcomatoid vs. epithelioid predominance was achieved, reducing subjective categorization.
  • The method demonstrated strong performance with an AUROC of 0.89±0.05 on a dataset of 234 tissue micro-array cores.

Conclusions:

  • The developed MIL approach provides an objective and accurate method for malignant mesothelioma subtyping.
  • This computational pathology technique can aid in understanding tumor heterogeneity and improving patient stratification.
  • The methodology and dataset are publicly available to advance research in mesothelioma diagnosis.