Intelligent Prediction of Pressure Injury by Image-Based Feature Variable With Machine Learning.

IF 2.2 4区 医学 Q1 NURSING
Xuehua Liu, Chengbin Tang, Lingxiang Guo, Jun Shao, Gang Wu, Yaru Qi
{"title":"Intelligent Prediction of Pressure Injury by Image-Based Feature Variable With Machine Learning.","authors":"Xuehua Liu, Chengbin Tang, Lingxiang Guo, Jun Shao, Gang Wu, Yaru Qi","doi":"10.1097/NNR.0000000000000846","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Rapid and objective pressure injury assessment is crucial for preventing further wound deterioration.</p><p><strong>Objectives: </strong>This study aimed to develop an image-based intelligent system for pressure injury (PI) determination that do not rely on human sensory evaluation.</p><p><strong>Methods: </strong>An image-based PI determination system was developed using a combination method of feature variable extraction and machine learning. Color and texture features were selected because they are closely related to human sensory evaluation methods. The digital data from these selected feature variables served as the original data set for model construction. Then, the contribution and relationships between the extracted feature variables and model performance were investigated using shapely additive explanations and Spearman algorithms to enhance the robustness of the PI determination model. Additionally, the influence of sample size and K values on model performance was determined for robust model construction.</p><p><strong>Results: </strong>A k-nearest neighbor algorithm was used to build pressure injury prediction models based on these selected variables and image samples. The classification rate for the best model is 97.22% and 97.08% on the training and test sets, respectively.</p><p><strong>Discussion: </strong>All results demonstrate that image-based feature variables coupled with machine learning are efficient for PI determination and perhaps other medical diagnoses involving visual recognition.</p>","PeriodicalId":49723,"journal":{"name":"Nursing Research","volume":" ","pages":""},"PeriodicalIF":2.2000,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nursing Research","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/NNR.0000000000000846","RegionNum":4,"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

Background: Rapid and objective pressure injury assessment is crucial for preventing further wound deterioration.

Objectives: This study aimed to develop an image-based intelligent system for pressure injury (PI) determination that do not rely on human sensory evaluation.

Methods: An image-based PI determination system was developed using a combination method of feature variable extraction and machine learning. Color and texture features were selected because they are closely related to human sensory evaluation methods. The digital data from these selected feature variables served as the original data set for model construction. Then, the contribution and relationships between the extracted feature variables and model performance were investigated using shapely additive explanations and Spearman algorithms to enhance the robustness of the PI determination model. Additionally, the influence of sample size and K values on model performance was determined for robust model construction.

Results: A k-nearest neighbor algorithm was used to build pressure injury prediction models based on these selected variables and image samples. The classification rate for the best model is 97.22% and 97.08% on the training and test sets, respectively.

Discussion: All results demonstrate that image-based feature variables coupled with machine learning are efficient for PI determination and perhaps other medical diagnoses involving visual recognition.

基于图像特征变量与机器学习的压力损伤智能预测。
背景:快速、客观的压力性损伤评估是防止伤口进一步恶化的关键。目的:本研究旨在开发一种不依赖于人类感官评估的基于图像的压力损伤(PI)智能检测系统。方法:采用特征变量提取与机器学习相结合的方法,建立基于图像的PI检测系统。选择颜色和纹理特征是因为它们与人类感官评价方法密切相关。从这些选定的特征变量中得到的数字数据作为模型构建的原始数据集。然后,使用形状加性解释和Spearman算法研究提取的特征变量与模型性能之间的贡献和关系,以增强PI确定模型的鲁棒性。此外,还确定了样本量和K值对模型性能的影响,以实现稳健的模型构建。结果:基于选取的变量和图像样本,采用k近邻算法建立压力损伤预测模型。最佳模型在训练集和测试集上的分类率分别为97.22%和97.08%。讨论:所有结果都表明,基于图像的特征变量与机器学习相结合,对于PI的确定和其他涉及视觉识别的医学诊断是有效的。
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 求助全文
来源期刊
Nursing Research
Nursing Research 医学-护理
CiteScore
3.60
自引率
4.00%
发文量
102
审稿时长
6-12 weeks
期刊介绍: Nursing Research is a peer-reviewed journal celebrating over 60 years as the most sought-after nursing resource; it offers more depth, more detail, and more of what today''s nurses demand. Nursing Research covers key issues, including health promotion, human responses to illness, acute care nursing research, symptom management, cost-effectiveness, vulnerable populations, health services, and community-based nursing studies. Each issue highlights the latest research techniques, quantitative and qualitative studies, and new state-of-the-art methodological strategies, including information not yet found in textbooks. Expert commentaries and briefs are also included. In addition to 6 issues per year, Nursing Research from time to time publishes supplemental content not found anywhere else.
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信