AI-Powered Image-Based Assessment of Pressure Injuries Using You Only Look once (YOLO) Version 8 Models.

IF 5.8 3区 医学 Q1 DERMATOLOGY Advances in wound care Pub Date : 2025-03-13 DOI:10.1089/wound.2024.0245
Mehedi Hasan Tusar, Fateme Fayyazbakhsh, Niloofar Zendehdel, Eduard Mochalin, Igor Melnychuk, Lisa Gould, Ming C Leu
{"title":"AI-Powered Image-Based Assessment of Pressure Injuries Using You Only Look once (YOLO) Version 8 Models.","authors":"Mehedi Hasan Tusar, Fateme Fayyazbakhsh, Niloofar Zendehdel, Eduard Mochalin, Igor Melnychuk, Lisa Gould, Ming C Leu","doi":"10.1089/wound.2024.0245","DOIUrl":null,"url":null,"abstract":"<p><p><b>Objective:</b> The primary objective of this study is to enhance the detection and staging of pressure injuries using machine learning capabilities for precise image analysis. This study explores the application of the You Only Look Once version 8 (YOLOv8) deep learning model for pressure injury staging. <b>Approach:</b> We prepared a high-quality, publicly available dataset to evaluate different variants of YOLOv8 (YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x) and five optimizers (Adam, AdamW, NAdam, RAdam, and stochastic gradient descent) to determine the most effective configuration. We followed a simulation-based research approach, which is an extension of the Consolidated Standards of Reporting Trials (CONSORT) and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for dataset preparation and algorithm evaluation. <b>Results:</b> YOLOv8s, with the AdamW optimizer and hyperparameter tuning, achieved the best performance metrics, including a mean average precision at intersection over union ≥0.5 of 84.16% and a recall of 82.31%, surpassing previous YOLO-based models in accuracy. The ensemble model incorporating all YOLOv8 variants showed strong performance when applied to unseen images. <b>Innovation:</b> Notably, the YOLOv8s model significantly improved detection for challenging stages such as Stage 2 and achieved accuracy rates of 0.90 for deep tissue injury, 0.91 for Unstageable, and 0.74, 0.76, 0.70, and 0.77 for Stages 1, 2, 3, and 4, respectively. <b>Conclusion:</b> These results demonstrate the effectiveness of YOLOv8s and ensemble models in improving the accuracy and robustness of pressure injury staging, offering a reliable tool for clinical decision-making.</p>","PeriodicalId":7413,"journal":{"name":"Advances in wound care","volume":" ","pages":""},"PeriodicalIF":5.8000,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advances in wound care","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1089/wound.2024.0245","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"DERMATOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Objective: The primary objective of this study is to enhance the detection and staging of pressure injuries using machine learning capabilities for precise image analysis. This study explores the application of the You Only Look Once version 8 (YOLOv8) deep learning model for pressure injury staging. Approach: We prepared a high-quality, publicly available dataset to evaluate different variants of YOLOv8 (YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x) and five optimizers (Adam, AdamW, NAdam, RAdam, and stochastic gradient descent) to determine the most effective configuration. We followed a simulation-based research approach, which is an extension of the Consolidated Standards of Reporting Trials (CONSORT) and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for dataset preparation and algorithm evaluation. Results: YOLOv8s, with the AdamW optimizer and hyperparameter tuning, achieved the best performance metrics, including a mean average precision at intersection over union ≥0.5 of 84.16% and a recall of 82.31%, surpassing previous YOLO-based models in accuracy. The ensemble model incorporating all YOLOv8 variants showed strong performance when applied to unseen images. Innovation: Notably, the YOLOv8s model significantly improved detection for challenging stages such as Stage 2 and achieved accuracy rates of 0.90 for deep tissue injury, 0.91 for Unstageable, and 0.74, 0.76, 0.70, and 0.77 for Stages 1, 2, 3, and 4, respectively. Conclusion: These results demonstrate the effectiveness of YOLOv8s and ensemble models in improving the accuracy and robustness of pressure injury staging, offering a reliable tool for clinical decision-making.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
求助全文
约1分钟内获得全文 去求助
来源期刊
Advances in wound care
Advances in wound care Medicine-Emergency Medicine
CiteScore
12.10
自引率
4.10%
发文量
62
期刊介绍: Advances in Wound Care rapidly shares research from bench to bedside, with wound care applications for burns, major trauma, blast injuries, surgery, and diabetic ulcers. The Journal provides a critical, peer-reviewed forum for the field of tissue injury and repair, with an emphasis on acute and chronic wounds. Advances in Wound Care explores novel research approaches and practices to deliver the latest scientific discoveries and developments. Advances in Wound Care coverage includes: Skin bioengineering, Skin and tissue regeneration, Acute, chronic, and complex wounds, Dressings, Anti-scar strategies, Inflammation, Burns and healing, Biofilm, Oxygen and angiogenesis, Critical limb ischemia, Military wound care, New devices and technologies.
期刊最新文献
AI-Powered Image-Based Assessment of Pressure Injuries Using You Only Look once (YOLO) Version 8 Models. Bridging the Gap: Contemporary Paradigms in Nerve Scaffolds and Regeneration. Flexible Pressure Sensors for Optimizing Pressure Garment Therapy in Periarticular Scar Treatment: Preclinical and Clinical Applications. Using Three-Dimensional Bioprinting to Generate Realistic Models of Wound Healing. Re: "A Novel Randomized Trial Protocol for Evaluating Wound Healing Interventions" by Bull et al.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1