Unsupervised machine learning to analyze corneal tissue surfaces

Carolin A. Rickert, Fabio Henkel, Oliver Lieleg
{"title":"Unsupervised machine learning to analyze corneal tissue surfaces","authors":"Carolin A. Rickert, Fabio Henkel, Oliver Lieleg","doi":"10.1063/5.0159502","DOIUrl":null,"url":null,"abstract":"Identifying/classifying damage features on soft materials, such as tissues, is much more challenging than on classical, hard materials—but nevertheless important, especially in the field of bio-tribology. For instance, cartilage samples from osteoarthritic patients exhibit surface damage even at early stages of tissue degeneration, and corneal tissues can be damaged by contact lenses when the ocular lubrication system fails. Here, we employ unsupervised machine learning (ML) methods to assess the surface condition of a soft tissue by detecting and classifying different wear morphologies as well as the severity of surface damage they represent. We show that different clustering methods, especially a k-means clustering algorithm, can indeed achieve a—from a material science point of view—meaningful classification of those tissue samples. Our study pinpoints the ability of unsupervised ML models to guide or even replace human decision processes for the analysis of complex surfaces and topographical datasets that—either owing to their complexity or the sample size—exceed the capability of the human brain.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"80 21","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"APL Machine Learning","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1063/5.0159502","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Identifying/classifying damage features on soft materials, such as tissues, is much more challenging than on classical, hard materials—but nevertheless important, especially in the field of bio-tribology. For instance, cartilage samples from osteoarthritic patients exhibit surface damage even at early stages of tissue degeneration, and corneal tissues can be damaged by contact lenses when the ocular lubrication system fails. Here, we employ unsupervised machine learning (ML) methods to assess the surface condition of a soft tissue by detecting and classifying different wear morphologies as well as the severity of surface damage they represent. We show that different clustering methods, especially a k-means clustering algorithm, can indeed achieve a—from a material science point of view—meaningful classification of those tissue samples. Our study pinpoints the ability of unsupervised ML models to guide or even replace human decision processes for the analysis of complex surfaces and topographical datasets that—either owing to their complexity or the sample size—exceed the capability of the human brain.
无监督机器学习分析角膜组织表面
识别/分类软材料(如组织)的损伤特征比传统的硬材料更具挑战性,但仍然很重要,特别是在生物摩擦学领域。例如,骨关节炎患者的软骨样本即使在组织退化的早期阶段也会出现表面损伤,当眼部润滑系统失效时,角膜组织可能会被隐形眼镜损伤。在这里,我们采用无监督机器学习(ML)方法,通过检测和分类不同的磨损形态以及它们所代表的表面损伤的严重程度来评估软组织的表面状况。我们表明,不同的聚类方法,特别是k-means聚类算法,确实可以从材料科学的角度对这些组织样本进行有意义的分类。我们的研究指出了无监督机器学习模型在复杂表面和地形数据集分析中指导甚至取代人类决策过程的能力,这些决策过程无论是由于其复杂性还是样本量,都超过了人类大脑的能力。
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
自引率
0.00%
发文量
0
×
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:481959085
Book学术官方微信