Detection of malignancy in whole slide images of endometrial cancer biopsies using artificial intelligence

Sep 24, 2026

This study uses artificial intelligence to categorize endometrial biopsy whole slide images from digital pathology as malignant, other or benign, or insufficient. Endometrial biopsy interpretation is a key step in diagnosis of endometrial cancer and is performed by pathologists. With digitization, slide availability enables automation through AI, which could prioritize cases for review and reduce time to diagnosis. The work trains a fully supervised CNN using 2909 pathologist-annotated slides, produces patch-level heatmaps, and then trains a slide classifier. The final model correctly classifies 90% of all slides and achieves 97% accuracy within the malignant class.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages28

Reading Time1 h 11 min

AccessFree to download

Abstract

Introduction

PLOS ONE OPEN ACCESS Citation: Fell C, Mohammadi M, Morrison D, Arandjelović O, Syed S, Konanahalli P, et al. (2023) Detection of malignancy in whole slide images of endometrial cancer biopsies using artificial intelligence. PLoS ONE 18(3): e0282577 . [https://](https://)[ ](https://)[doi.org/10.1371/journal.pone.0282577](doi.org/10.1371/journal.pone.0282577) Editor: Vincenzo L’Imperio, Universita degli Studi di Milano-Bicocca, ITALY Received: November 7, 2022 Accepted: February 21, 2023 Published: March 8, 2023 Copyright: © 2023 Fell et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: Data cannot be shared publicly because it is patient data and the approvals do not permit us to release the data. The data underlying the results and ethics approval for use was obtained from the NHS greater Glasgow and Clyde Biorepository and Pathology Tissue Resource ([https://www.nhsgbr.org.uk](https://www.nhsgbr.org.uk)) . Funding: For all authors this work is supported by the Industrial Centre for AI Research in digital Diagnostics (iCAIRD) which is funded by Innovate UK ([https://www.ukri.org/councils/innovate-uk/](https://www.ukri.org/councils/innovate-uk/)) on RESEARCH ARTICLE Detection of malignancy in whole slide images of endometrial cancer biopsies using artificial intelligence Christina Fell1 *, Mahnaz Mohammadi1, David Morrison1, Ognjen Arandjelović1, Sheeba Syed2, Prakash Konanahalli2, Sarah Bell2, Gareth Bryson2, David J. Harrison3,4, David Harris-Birtill1 1 School of Computer Science, University of St Andrews, St Andrews, Scotland, United Kingdom, 2 Pathology Department, NHS Greater Glasgow and Clyde, Glasgow, Scotland, United Kingdom, 3 School of Medicine, University of St Andrews, St Andrews, Scotland, United Kingdom, 4 NHS Lothian Pathology, Division of Laboratory Medicine, Royal Infirmary of Edinburgh, Edinburgh, Scotland, United Kingdom * [cmf21@st-andrews.ac.uk](cmf21@st-andrews.ac.uk) Abstract In this study we use artificial intelligence (AI) to categorise endometrial biopsy whole slide images (WSI) from digital pathology as either “malignant”,“other or benign” or “insufficient”. An endometrial biopsy is a key step in diagnosis of endometrial cancer, biopsies are viewed and diagnosed by pathologists. Pathology is increasingly digitised, with slides viewed as images on screens rather than through the lens of a microscope. The availability of these images is driving automation via the application of AI. A model that classifies slides in the manner proposed would allow prioritisation of these slides for pathologist review and hence reduce time to diagnosis for patients with cancer. Previous studies using AI on endometrial biopsies have examined slightly different tasks, for example using images alongside genomic data to differentiate between cancer subtypes. We took 2909 slides with “malignant”and “other or benign” areas annotated by pathologists. A fully supervised convolutional neural network (CNN) model was trained to calculate the probability of a patch from the slide being “malignant” or “other or benign”. Heatmaps of all the patches on each slide were then produced to show malignant areas. These heatmaps were used to train a slide classification model to give the final slide categorisation as either “malignant”,“other or benign” or “insufficient”. The final model was able to accurately classify 90% of all slides correctly and 97% of slides in the malignant class; this accuracy is good enough to allow prioritisation of pathologists’ workload. 1 Introduction The aim of this work is to automatically sort histopathology whole slide images (WSI) of endometrial biopsies into one of three categories,“malignant”,“other or benign” or “insufficient”. This would allow prioritisa

FAQs

What categories does the AI model classify for endometrial biopsy whole slide images?

The model categorizes slides into three groups: malignant, other or benign, or insufficient.

How was the model trained and how are malignant regions identified?

A fully supervised convolutional neural network was trained on patch-level labels from 2909 pathologist-annotated slides. Heatmaps of malignant areas across patches were generated and used to train the final slide classification model.

What classification performance does the study report?

The final model correctly classifies 90% of all slides overall and reaches 97% accuracy for slides in the malignant class.

Sep 24, 2026

Uploaded by MrHarris58