机器学习在静脉血栓栓塞中的应用——为什么?下一步是什么?

IF 4.1 2区 医学 Q2 HEMATOLOGY
Gerard Gurumurthy, Filip Kisiel, Lianna Reynolds, Will Thomas, Maha Othman, Deepa J Arachchillage, Jecko Thachil
{"title":"机器学习在静脉血栓栓塞中的应用——为什么?下一步是什么?","authors":"Gerard Gurumurthy, Filip Kisiel, Lianna Reynolds, Will Thomas, Maha Othman, Deepa J Arachchillage, Jecko Thachil","doi":"10.1055/a-2669-7933","DOIUrl":null,"url":null,"abstract":"<p><p>Venous thromboembolism (VTE) remains a leading cause of cardiovascular morbidity and mortality, despite advances in imaging and anticoagulation. VTE arises from diverse and overlapping risk factors, such as inherited thrombophilia, immobility, malignancy, surgery or trauma, pregnancy, hormonal therapy, obesity, chronic medical conditions (e.g., heart failure, inflammatory disease), and advancing age. Clinicians, therefore, face challenges in balancing the benefits of thromboprophylaxis against the bleeding risk. Existing clinical risk scores often exhibit only modest discrimination and calibration across heterogeneous patient populations. Machine learning (ML) has emerged as a promising tool to address these limitations. In imaging, convolutional neural networks and hybrid algorithms can detect VTE on CT pulmonary angiography with areas under the curves (AUCs) of 0.85 to 0.96. In surgical cohorts, gradient-boosting models outperform traditional risk scores, achieving AUCs between 0.70 and 0.80 in predicting postoperative VTE. In cancer-associated venous thrombosis, advanced ML models demonstrate AUCs between 0.68 and 0.82. However, concerns about bias and external validation persist. Bleeding risk prediction models remain challenging in extended anticoagulation settings, often matching conventional models. Predicting recurrent VTE using neural networks showed AUCs of 0.93 to 0.99 in initial studies. However, these lack transparency and prospective validation. Most ML models suffer from limited external validation, \"black box\" algorithms, and integration hurdles within clinical workflows. Future efforts should focus on standardized reporting (e.g., Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis [TRIPOD]-ML), transparent model interpretation, prospective impact assessments, and seamless incorporation into electronic health records to realize the full potential of ML in VTE.</p>","PeriodicalId":21673,"journal":{"name":"Seminars in thrombosis and hemostasis","volume":" ","pages":""},"PeriodicalIF":4.1000,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Machine Learning in Venous Thromboembolism - Why and What Next?\",\"authors\":\"Gerard Gurumurthy, Filip Kisiel, Lianna Reynolds, Will Thomas, Maha Othman, Deepa J Arachchillage, Jecko Thachil\",\"doi\":\"10.1055/a-2669-7933\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Venous thromboembolism (VTE) remains a leading cause of cardiovascular morbidity and mortality, despite advances in imaging and anticoagulation. VTE arises from diverse and overlapping risk factors, such as inherited thrombophilia, immobility, malignancy, surgery or trauma, pregnancy, hormonal therapy, obesity, chronic medical conditions (e.g., heart failure, inflammatory disease), and advancing age. Clinicians, therefore, face challenges in balancing the benefits of thromboprophylaxis against the bleeding risk. Existing clinical risk scores often exhibit only modest discrimination and calibration across heterogeneous patient populations. Machine learning (ML) has emerged as a promising tool to address these limitations. In imaging, convolutional neural networks and hybrid algorithms can detect VTE on CT pulmonary angiography with areas under the curves (AUCs) of 0.85 to 0.96. In surgical cohorts, gradient-boosting models outperform traditional risk scores, achieving AUCs between 0.70 and 0.80 in predicting postoperative VTE. In cancer-associated venous thrombosis, advanced ML models demonstrate AUCs between 0.68 and 0.82. However, concerns about bias and external validation persist. Bleeding risk prediction models remain challenging in extended anticoagulation settings, often matching conventional models. Predicting recurrent VTE using neural networks showed AUCs of 0.93 to 0.99 in initial studies. However, these lack transparency and prospective validation. Most ML models suffer from limited external validation, \\\"black box\\\" algorithms, and integration hurdles within clinical workflows. Future efforts should focus on standardized reporting (e.g., Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis [TRIPOD]-ML), transparent model interpretation, prospective impact assessments, and seamless incorporation into electronic health records to realize the full potential of ML in VTE.</p>\",\"PeriodicalId\":21673,\"journal\":{\"name\":\"Seminars in thrombosis and hemostasis\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":4.1000,\"publicationDate\":\"2025-08-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Seminars in thrombosis and hemostasis\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1055/a-2669-7933\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"HEMATOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Seminars in thrombosis and hemostasis","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1055/a-2669-7933","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"HEMATOLOGY","Score":null,"Total":0}
引用次数: 0

摘要

静脉血栓栓塞(VTE)仍然是心血管疾病发病率和死亡率的主要原因,尽管在成像和抗凝方面取得了进展。静脉血栓栓塞由多种重叠的危险因素引起,如遗传性血栓形成、不活动、恶性肿瘤、手术或创伤、妊娠、激素治疗、肥胖、慢性疾病(如心力衰竭、炎症性疾病)和年龄增长。因此,临床医生在平衡血栓预防的益处和出血风险方面面临挑战。现有的临床风险评分在异质患者群体中往往只表现出适度的区分和校准。机器学习(ML)已经成为解决这些限制的有前途的工具。在影像学上,卷积神经网络和混合算法可以在CT肺血管造影上检测到VTE,曲线下面积(aus)为0.85 ~ 0.96。在手术队列中,梯度增强模型优于传统的风险评分,预测术后静脉血栓栓塞的auc在0.70至0.80之间。在癌症相关静脉血栓中,晚期ML模型显示auc在0.68至0.82之间。然而,对偏倚和外部验证的担忧仍然存在。出血风险预测模型在扩展抗凝设置中仍然具有挑战性,通常与传统模型相匹配。在最初的研究中,使用神经网络预测静脉血栓栓塞复发的auc为0.93 ~ 0.99。然而,这些缺乏透明度和前瞻性验证。大多数机器学习模型都受到有限的外部验证、“黑盒”算法和临床工作流程中的集成障碍的影响。未来的工作应侧重于标准化报告(例如,透明报告个体预后或诊断的多变量预测模型[TRIPOD]-ML),透明模型解释,前瞻性影响评估,以及无缝整合到电子健康记录中,以实现VTE中ML的全部潜力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Machine Learning in Venous Thromboembolism - Why and What Next?

Venous thromboembolism (VTE) remains a leading cause of cardiovascular morbidity and mortality, despite advances in imaging and anticoagulation. VTE arises from diverse and overlapping risk factors, such as inherited thrombophilia, immobility, malignancy, surgery or trauma, pregnancy, hormonal therapy, obesity, chronic medical conditions (e.g., heart failure, inflammatory disease), and advancing age. Clinicians, therefore, face challenges in balancing the benefits of thromboprophylaxis against the bleeding risk. Existing clinical risk scores often exhibit only modest discrimination and calibration across heterogeneous patient populations. Machine learning (ML) has emerged as a promising tool to address these limitations. In imaging, convolutional neural networks and hybrid algorithms can detect VTE on CT pulmonary angiography with areas under the curves (AUCs) of 0.85 to 0.96. In surgical cohorts, gradient-boosting models outperform traditional risk scores, achieving AUCs between 0.70 and 0.80 in predicting postoperative VTE. In cancer-associated venous thrombosis, advanced ML models demonstrate AUCs between 0.68 and 0.82. However, concerns about bias and external validation persist. Bleeding risk prediction models remain challenging in extended anticoagulation settings, often matching conventional models. Predicting recurrent VTE using neural networks showed AUCs of 0.93 to 0.99 in initial studies. However, these lack transparency and prospective validation. Most ML models suffer from limited external validation, "black box" algorithms, and integration hurdles within clinical workflows. Future efforts should focus on standardized reporting (e.g., Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis [TRIPOD]-ML), transparent model interpretation, prospective impact assessments, and seamless incorporation into electronic health records to realize the full potential of ML in VTE.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Seminars in thrombosis and hemostasis
Seminars in thrombosis and hemostasis 医学-外周血管病
CiteScore
8.80
自引率
21.10%
发文量
132
审稿时长
6-12 weeks
期刊介绍: Seminars in Thrombosis and Hemostasis is a topic driven review journal that focuses on all issues relating to hemostatic and thrombotic disorders. As one of the premiere review journals in the field, Seminars in Thrombosis and Hemostasis serves as a comprehensive forum for important advances in clinical and laboratory diagnosis and therapeutic interventions. The journal also publishes peer reviewed original research papers. Seminars offers an informed perspective on today''s pivotal issues, including hemophilia A & B, thrombophilia, gene therapy, venous and arterial thrombosis, von Willebrand disease, vascular disorders and thromboembolic diseases. Attention is also given to the latest developments in pharmaceutical drugs along with treatment and current management techniques. The journal also frequently publishes sponsored supplements to further highlight emerging trends in the field.
×
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