Mpox lesion counting with semantic and instance segmentation methods
Authors:
- Bohan JiangBJBohan Jiang
- Andrew J. McNeilAMAndrew J. McNeil
- Yihao LiuYLYihao Liu
- David W. HouseDHDavid W. House
- Placide Mbala-KingebeniPMPlacide Mbala-Kingebeni
- Olivier Tshiani MbayaOMOlivier Tshiani Mbaya
- Tyra SilaphetTSTyra Silaphet
- Lori E. DoddLDLori E. Dodd
- Edward W. CowenECEdward W. Cowen
- Veronique NussenblattVNVeronique Nussenblatt
- Tyler BonnettTBTyler Bonnett
- Ziche ChenZCZiche Chen
Open Access
- 1 May 2025
- journal article
- Published by SPIE-Intl Soc Optical Eng in Journal of Medical Imaging
- Vol. 12 (03) , 1-10
Abstract
PurposeMpox is a viral illness with symptoms similar to smallpox. A key clinical metric to monitor disease progression is the number of skin lesions. Manually counting mpox skin lesions is labor-intensive and susceptible to human error.ApproachWe previously developed an mpox lesion counting method based on the UNet segmentation model using 66 photographs from 18 patients. We have compared four additional methods: the instance segmentation methods Mask R-CNN, YOLOv8, and E2EC, in addition to a UNet++ model. We designed a patient-level leave-one-out experiment, assessing their performance using F1 score and lesion count metrics. Finally, we tested whether an ensemble of the networks outperformed any single model.ResultsMask R-CNN model achieved an F1 score of 0.75, YOLOv8 a score of 0.75, E2EC a score of 0.70, UNet++ a score of 0.81, and baseline UNet a score of 0.79. Bland-Altman analysis of lesion count performance showed a limit of agreement (LoA) width of 62.2 for Mask R-CNN, 91.3 for YOLOv8, 94.2 for E2EC, and 62.1 for UNet++, with the baseline UNet model achieving 69.1. The ensemble showed an F1 score performance of 0.78 and LoA width of 67.4.ConclusionsInstance segmentation methods and UNet-based semantic segmentation methods performed equally well in lesion counting. Furthermore, the ensemble of the trained models showed no performance increase over the best-performing model UNet, likely because errors are frequently shared across models. Performance is likely limited by the availability of high-quality photographs for this complex problem, rather than the methodologies used.Keywords
Funding Information
- National Institutes of Health (R01 R01HL169944)
- National Cancer Institute, National Institutes of Health (75N91019D00024)
- United States Department of Veterans Affairs Clinical Sciences R&D Service (IK2 CX001785)
- National Center for Advancing Translational Sciences of the National Institutes of Health (TL1TR002244)
This publication has 14 references indexed in Scilit:
- Counting Monkeypox Lesions in Patient Photographs: Limits of Agreement of Manual Counts and Artificial IntelligenceJournal of Investigative Dermatology, 2022
- Clinical characterization of human monkeypox infections in the Democratic Republic of the CongomedRxiv, 2022
- Artificial intelligence recognition of cutaneous chronic graft‐versus‐host disease by a deep learning neural networkBritish Journal of Haematology, 2022
- UNet++: A Nested U-Net Architecture for Medical Image Segmentation2018
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- You Only Look Once: Unified, Real-Time Object Detection2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Object Detection with Discriminatively Trained Part-Based ModelsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2009
- Reemergence of Monkeypox: Prevalence, Diagnostics, and CountermeasuresClinical Infectious Diseases, 2005
- STATISTICAL METHODS FOR ASSESSING AGREEMENT BETWEEN TWO METHODS OF CLINICAL MEASUREMENTThe Lancet, 1986