Development of a Deep Learning-Based Model for Pressure Injury Surface Assessment.

IF 3.2 3区医学 Q1 NURSING Journal of Clinical Nursing Pub Date : 2025-01-14 DOI:10.1111/jocn.17645

Ankang Liu, Hualong Ma, Yanying Zhu, Qinyang Wu, Shihai Xu, Wei Feng, Haobin Liang, Jian Ma, Xinwei Wang, Xuemei Ye, Yanxiong Liu, Chao Wang, Xu Sun, Shijun Xiang, Qiaohong Yang

{"title":"Development of a Deep Learning-Based Model for Pressure Injury Surface Assessment.","authors":"Ankang Liu, Hualong Ma, Yanying Zhu, Qinyang Wu, Shihai Xu, Wei Feng, Haobin Liang, Jian Ma, Xinwei Wang, Xuemei Ye, Yanxiong Liu, Chao Wang, Xu Sun, Shijun Xiang, Qiaohong Yang","doi":"10.1111/jocn.17645","DOIUrl":null,"url":null,"abstract":"Aim: To develop a deep learning-based smart assessment model for pressure injury surface.Design: Exploratory analysis study.Methods: Pressure injury images from four Guangzhou hospitals were labelled and used to train a neural network model. Evaluation metrics included mean intersection over union (MIoU), pixel accuracy (PA), and accuracy. Model performance was tested by comparing wound number, maximum dimensions and area extent.Results: From 1063 images, the model achieved 74% IoU, 88% PA and 83% accuracy for wound bed segmentation. Cohen's kappa coefficient for wound number was 0.810. Correlation coefficients were 0.900 for maximum length (mean difference 0.068 cm), 0.814 for maximum width (mean difference 0.108 cm) and 0.930 for regional extent (mean difference 0.527 cm2).Conclusion: The model demonstrated exceptional automated estimation capabilities, potentially serving as a crucial tool for informed decision-making in wound assessment.Implications and impact: This study promotes precision nursing and equitable resource use. The AI-based assessment model serves clinical work by assisting healthcare professionals in decision-making and facilitating wound assessment resource sharing.Reporting method: The STROBE checklist guided study reporting.Patient or public contribution: Patients provided image resources for model training.","PeriodicalId":50236,"journal":{"name":"Journal of Clinical Nursing","volume":" ","pages":""},"PeriodicalIF":3.2000,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Clinical Nursing","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1111/jocn.17645","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"NURSING","Score":null,"Total":0}

引用次数: 0

Abstract

Aim: To develop a deep learning-based smart assessment model for pressure injury surface.

Design: Exploratory analysis study.

Methods: Pressure injury images from four Guangzhou hospitals were labelled and used to train a neural network model. Evaluation metrics included mean intersection over union (MIoU), pixel accuracy (PA), and accuracy. Model performance was tested by comparing wound number, maximum dimensions and area extent.

Results: From 1063 images, the model achieved 74% IoU, 88% PA and 83% accuracy for wound bed segmentation. Cohen's kappa coefficient for wound number was 0.810. Correlation coefficients were 0.900 for maximum length (mean difference 0.068 cm), 0.814 for maximum width (mean difference 0.108 cm) and 0.930 for regional extent (mean difference 0.527 cm²).

Conclusion: The model demonstrated exceptional automated estimation capabilities, potentially serving as a crucial tool for informed decision-making in wound assessment.

Implications and impact: This study promotes precision nursing and equitable resource use. The AI-based assessment model serves clinical work by assisting healthcare professionals in decision-making and facilitating wound assessment resource sharing.

Reporting method: The STROBE checklist guided study reporting.

Patient or public contribution: Patients provided image resources for model training.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

基于深度学习的压力损伤面评估模型的开发。

目的：建立基于深度学习的压力损伤表面智能评估模型。设计：探索性分析研究。方法：对广州四家医院的压力损伤图像进行标记，并使用神经网络模型进行训练。评估指标包括平均交联（MIoU），像素精度（PA）和精度。通过比较伤口数量、最大尺寸和面积范围来检验模型的性能。结果：在1063张图像中，该模型对伤口床的分割达到了74%的IoU、88%的PA和83%的准确率。伤口数的Cohen’s kappa系数为0.810。最大长度相关系数为0.900（平均差值0.068 cm），最大宽度相关系数为0.814（平均差值0.108 cm），区域范围相关系数为0.930（平均差值0.527 cm2）。结论：该模型展示了卓越的自动估计能力，可能作为伤口评估中知情决策的重要工具。启示与影响：本研究促进精准护理和资源公平利用。基于人工智能的评估模型通过协助医疗专业人员决策和促进伤口评估资源共享来服务于临床工作。报告方法：STROBE检查表指导研究报告。患者或公众贡献：患者为模型训练提供图像资源。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Journal of Clinical Nursing 医学-护理

CiteScore

6.40

自引率

2.40%

发文量

审稿时长

2 months

期刊介绍： The Journal of Clinical Nursing (JCN) is an international, peer reviewed, scientific journal that seeks to promote the development and exchange of knowledge that is directly relevant to all spheres of nursing practice. The primary aim is to promote a high standard of clinically related scholarship which advances and supports the practice and discipline of nursing. The Journal also aims to promote the international exchange of ideas and experience that draws from the different cultures in which practice takes place. Further, JCN seeks to enrich insight into clinical need and the implications for nursing intervention and models of service delivery. Emphasis is placed on promoting critical debate on the art and science of nursing practice. JCN is essential reading for anyone involved in nursing practice, whether clinicians, researchers, educators, managers, policy makers, or students. The development of clinical practice and the changing patterns of inter-professional working are also central to JCN''s scope of interest. Contributions are welcomed from other health professionals on issues that have a direct impact on nursing practice. We publish high quality papers from across the methodological spectrum that make an important and novel contribution to the field of clinical nursing (regardless of where care is provided), and which demonstrate clinical application and international relevance.