Ultrasound Radiomics and Dual-Mode Ultrasonic Elastography Based Machine Learning Model for the Classification of Benign and Malignant Thyroid Nodules.

IF 1.2 4区 医学 Q3 ACOUSTICS
Junhong Yan, Xuemin Zhou, Qi Zheng, Kun Wang, Yanbing Gao, Feifei Liu, Lei Pan
{"title":"Ultrasound Radiomics and Dual-Mode Ultrasonic Elastography Based Machine Learning Model for the Classification of Benign and Malignant Thyroid Nodules.","authors":"Junhong Yan, Xuemin Zhou, Qi Zheng, Kun Wang, Yanbing Gao, Feifei Liu, Lei Pan","doi":"10.1002/jcu.24104","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>The present study aims to construct a random forest (RF) model based on ultrasound radiomics and elastography, offering a new approach for the differentiation of thyroid nodules (TNs).</p><p><strong>Methods: </strong>We retrospectively analyzed 152 TNs from 127 patients and developed four machine learning models. The examination was performed using the Resona 9Pro equipped with a 15-4 MHz linear array probe. The region of interest (ROI) was delineated with 3D Slicer. Using the RF algorithm, four models were developed based on sound touch elastography (STE) parameters, strain elastography (SE) parameters, and the selected radiomic features: the STE model, SE model, radiomics model, and the combined model. Decision Curve Analysis (DCA) is employed to assess the clinical benefit of each model. The DeLong test is used to determine whether the area under the curves (AUC) values of different models are statistically significant.</p><p><strong>Results: </strong>A total of 1396 radiomic features were extracted using the Pyradiomics package. After screening, a total of 7 radiomic features were ultimately included in the construction of the model. In STE, SE, radiomics model, and combined model, the AUCs are 0.699 (95% CI: 0.570-0.828), 0.812 (95% CI: 0.683-0.941), 0.851 (95% CI: 0.739-0.964) and 0.911 (95% CI: 0.806-1.000), respectively. In these models, the combined model and the radiomics model exhibited outstanding performance.</p><p><strong>Conclusions: </strong>The combined model, integrating elastography and radiomics, demonstrates superior predictive accuracy compared to single models, offering a promising approach for the diagnosis of TNs.</p>","PeriodicalId":15386,"journal":{"name":"Journal of Clinical Ultrasound","volume":" ","pages":""},"PeriodicalIF":1.2000,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Clinical Ultrasound","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1002/jcu.24104","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ACOUSTICS","Score":null,"Total":0}
引用次数: 0

Abstract

Introduction: The present study aims to construct a random forest (RF) model based on ultrasound radiomics and elastography, offering a new approach for the differentiation of thyroid nodules (TNs).

Methods: We retrospectively analyzed 152 TNs from 127 patients and developed four machine learning models. The examination was performed using the Resona 9Pro equipped with a 15-4 MHz linear array probe. The region of interest (ROI) was delineated with 3D Slicer. Using the RF algorithm, four models were developed based on sound touch elastography (STE) parameters, strain elastography (SE) parameters, and the selected radiomic features: the STE model, SE model, radiomics model, and the combined model. Decision Curve Analysis (DCA) is employed to assess the clinical benefit of each model. The DeLong test is used to determine whether the area under the curves (AUC) values of different models are statistically significant.

Results: A total of 1396 radiomic features were extracted using the Pyradiomics package. After screening, a total of 7 radiomic features were ultimately included in the construction of the model. In STE, SE, radiomics model, and combined model, the AUCs are 0.699 (95% CI: 0.570-0.828), 0.812 (95% CI: 0.683-0.941), 0.851 (95% CI: 0.739-0.964) and 0.911 (95% CI: 0.806-1.000), respectively. In these models, the combined model and the radiomics model exhibited outstanding performance.

Conclusions: The combined model, integrating elastography and radiomics, demonstrates superior predictive accuracy compared to single models, offering a promising approach for the diagnosis of TNs.

基于超声放射组学和双模超声弹性成像的良恶性甲状腺结节分类机器学习模型。
摘要:本研究旨在建立基于超声放射组学和弹性成像的随机森林(RF)模型,为甲状腺结节(TNs)的鉴别提供一种新的方法。方法:我们回顾性分析了127例患者的152例TNs,并建立了四种机器学习模型。使用配备15-4 MHz线性阵列探头的Resona 9Pro进行检查。利用三维切片器对感兴趣区域(ROI)进行了划分。利用射频算法,基于声触弹性成像(STE)参数、应变弹性成像(SE)参数和选定的放射组学特征建立了4个模型:STE模型、SE模型、放射组学模型和组合模型。采用决策曲线分析法(Decision Curve Analysis, DCA)评价各模型的临床疗效。采用DeLong检验确定不同模型的曲线下面积(AUC)值是否具有统计学显著性。结果:利用Pyradiomics包提取了1396个放射组学特征。经过筛选,最终共有7个放射学特征被纳入模型构建。STE、SE、放射组学模型和联合模型的auc分别为0.699 (95% CI: 0.570 ~ 0.828)、0.812 (95% CI: 0.683 ~ 0.941)、0.851 (95% CI: 0.739 ~ 0.964)和0.911 (95% CI: 0.806 ~ 1.000)。在这些模型中,组合模型和放射组学模型表现出突出的性能。结论:与单一模型相比,结合弹性成像和放射组学的联合模型具有更高的预测准确性,为TNs的诊断提供了一种有希望的方法。
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
CiteScore
1.90
自引率
0.00%
发文量
248
审稿时长
6 months
期刊介绍: The Journal of Clinical Ultrasound (JCU) is an international journal dedicated to the worldwide dissemination of scientific information on diagnostic and therapeutic applications of medical sonography. The scope of the journal includes--but is not limited to--the following areas: sonography of the gastrointestinal tract, genitourinary tract, vascular system, nervous system, head and neck, chest, breast, musculoskeletal system, and other superficial structures; Doppler applications; obstetric and pediatric applications; and interventional sonography. Studies comparing sonography with other imaging modalities are encouraged, as are studies evaluating the economic impact of sonography. Also within the journal''s scope are innovations and improvements in instrumentation and examination techniques and the use of contrast agents. JCU publishes original research articles, case reports, pictorial essays, technical notes, and letters to the editor. The journal is also dedicated to being an educational resource for its readers, through the publication of review articles and various scientific contributions from members of the editorial board and other world-renowned experts in sonography.
×
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