Analysis of nailfold capillaroscopy images with artificial intelligence: Data from literature and performance of machine learning and deep learning from images acquired in the SCLEROCAP study

IF 2.9 4区 医学 Q2 PERIPHERAL VASCULAR DISEASE
Lutfi Ozturk , Charlotte Laclau , Carine Boulon , Marion Mangin , Etheve Braz-ma , Joel Constans , Loubna Dari , Claire Le Hello
{"title":"Analysis of nailfold capillaroscopy images with artificial intelligence: Data from literature and performance of machine learning and deep learning from images acquired in the SCLEROCAP study","authors":"Lutfi Ozturk ,&nbsp;Charlotte Laclau ,&nbsp;Carine Boulon ,&nbsp;Marion Mangin ,&nbsp;Etheve Braz-ma ,&nbsp;Joel Constans ,&nbsp;Loubna Dari ,&nbsp;Claire Le Hello","doi":"10.1016/j.mvr.2024.104753","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><div>To evaluate the performance of machine learning and then deep learning to detect a systemic scleroderma (SSc) landscape from the same set of nailfold capillaroscopy (NC) images from the French prospective multicenter observational study SCLEROCAP.</div></div><div><h3>Methods</h3><div>NC images from the first 100 SCLEROCAP patients were analyzed to assess the performance of machine learning and then deep learning in identifying the SSc landscape, the NC images having previously been independently and consensually labeled by expert clinicians. Images were divided into a training set (70 %) and a validation set (30 %). After features extraction from the NC images, we tested six classifiers (random forests (RF), support vector machine (SVM), logistic regression (LR), light gradient boosting (LGB), extreme gradient boosting (XGB), K-nearest neighbors (KNN)) on the training set with five different combinations of the images. The performance of each classifier was evaluated by the F1 score. In the deep learning section, we tested three pre-trained models from the TIMM library (ResNet-18, DenseNet-121 and VGG-16) on raw NC images after applying image augmentation methods.</div></div><div><h3>Results</h3><div>With machine learning, performance ranged from 0.60 to 0.73 for each variable, with Hu and Haralick moments being the most discriminating. Performance was highest with the RF, LGB and XGB models (F1 scores: 0.75–0.79). The highest score was obtained by combining all variables and using the LGB model (F1 score: 0.79 ± 0.05, <em>p</em> &lt; 0.01). With deep learning, performance reached a minimum accuracy of 0.87. The best results were obtained with the DenseNet-121 model (accuracy 0.94 ± 0.02, F1 score 0.94 ± 0.02, AUC 0.95 ± 0.03) as compared to ResNet-18 (accuracy 0.87 ± 0.04, F1 score 0.85 ± 0.03, AUC 0.87 ± 0.04) and VGG-16 (accuracy 0.90 ± 0.03, F1 score 0.91 ± 0.02, AUC 0.91 ± 0.04).</div></div><div><h3>Conclusion</h3><div>By using machine learning and then deep learning on the same set of labeled NC images from the SCLEROCAP study, the highest performances to detect SSc landscape were obtained with deep learning and in particular DenseNet-121. This pre-trained model could therefore be used to automatically interpret NC images in case of suspected SSc. This result nevertheless needs to be confirmed on a larger number of NC images.</div></div>","PeriodicalId":18534,"journal":{"name":"Microvascular research","volume":"157 ","pages":"Article 104753"},"PeriodicalIF":2.9000,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Microvascular research","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S002628622400102X","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PERIPHERAL VASCULAR DISEASE","Score":null,"Total":0}
引用次数: 0

Abstract

Objective

To evaluate the performance of machine learning and then deep learning to detect a systemic scleroderma (SSc) landscape from the same set of nailfold capillaroscopy (NC) images from the French prospective multicenter observational study SCLEROCAP.

Methods

NC images from the first 100 SCLEROCAP patients were analyzed to assess the performance of machine learning and then deep learning in identifying the SSc landscape, the NC images having previously been independently and consensually labeled by expert clinicians. Images were divided into a training set (70 %) and a validation set (30 %). After features extraction from the NC images, we tested six classifiers (random forests (RF), support vector machine (SVM), logistic regression (LR), light gradient boosting (LGB), extreme gradient boosting (XGB), K-nearest neighbors (KNN)) on the training set with five different combinations of the images. The performance of each classifier was evaluated by the F1 score. In the deep learning section, we tested three pre-trained models from the TIMM library (ResNet-18, DenseNet-121 and VGG-16) on raw NC images after applying image augmentation methods.

Results

With machine learning, performance ranged from 0.60 to 0.73 for each variable, with Hu and Haralick moments being the most discriminating. Performance was highest with the RF, LGB and XGB models (F1 scores: 0.75–0.79). The highest score was obtained by combining all variables and using the LGB model (F1 score: 0.79 ± 0.05, p < 0.01). With deep learning, performance reached a minimum accuracy of 0.87. The best results were obtained with the DenseNet-121 model (accuracy 0.94 ± 0.02, F1 score 0.94 ± 0.02, AUC 0.95 ± 0.03) as compared to ResNet-18 (accuracy 0.87 ± 0.04, F1 score 0.85 ± 0.03, AUC 0.87 ± 0.04) and VGG-16 (accuracy 0.90 ± 0.03, F1 score 0.91 ± 0.02, AUC 0.91 ± 0.04).

Conclusion

By using machine learning and then deep learning on the same set of labeled NC images from the SCLEROCAP study, the highest performances to detect SSc landscape were obtained with deep learning and in particular DenseNet-121. This pre-trained model could therefore be used to automatically interpret NC images in case of suspected SSc. This result nevertheless needs to be confirmed on a larger number of NC images.
用人工智能分析甲襞毛细血管镜图像:文献数据以及 SCLEROCAP 研究中获取的图像的机器学习和深度学习性能。
目的评估机器学习和深度学习从法国前瞻性多中心观察研究SCLEROCAP的同一组甲襞毛细血管镜(NC)图像中检测系统性硬皮病(SSc)景观的性能:对 SCLEROCAP 首批 100 名患者的 NC 图像进行了分析,以评估机器学习和深度学习在识别系统性红斑狼疮景观方面的性能。图像分为训练集(70%)和验证集(30%)。从 NC 图像中提取特征后,我们在训练集上用五种不同的图像组合测试了六种分类器(随机森林 (RF)、支持向量机 (SVM)、逻辑回归 (LR)、轻梯度提升 (LGB)、极度梯度提升 (XGB)、K-近邻 (KNN))。每个分类器的性能通过 F1 分数进行评估。在深度学习部分,我们在应用图像增强方法后的原始数控图像上测试了 TIMM 库中的三个预训练模型(ResNet-18、DenseNet-121 和 VGG-16):通过机器学习,每个变量的性能从 0.60 到 0.73 不等,其中 Hu 矩和 Haralick 矩的判别能力最强。RF、LGB 和 XGB 模型的性能最高(F1 分数:0.75-0.79)。综合所有变量并使用 LGB 模型的得分最高(F1 得分:0.79 ± 0.05,p 结论:LGB 模型的得分最高:通过对 SCLEROCAP 研究中的同一组标注 NC 图像进行机器学习和深度学习,深度学习,特别是 DenseNet-121 获得了检测 SSc 景观的最高分。因此,这种预训练模型可用于自动解读疑似 SSc 的 NC 图像。不过,这一结果还需要在更多的 NC 图像上得到证实。
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
Microvascular research
Microvascular research 医学-外周血管病
CiteScore
6.00
自引率
3.20%
发文量
158
审稿时长
43 days
期刊介绍: Microvascular Research is dedicated to the dissemination of fundamental information related to the microvascular field. Full-length articles presenting the results of original research and brief communications are featured. Research Areas include: • Angiogenesis • Biochemistry • Bioengineering • Biomathematics • Biophysics • Cancer • Circulatory homeostasis • Comparative physiology • Drug delivery • Neuropharmacology • Microvascular pathology • Rheology • Tissue Engineering.
×
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