开发一种机器学习模型,用于在接受抗凝治疗的癌症相关血栓症患者中预测出血
Aleksandra G Grdinic1, Sandro Radovanovic2, Jostein Gleditsch3
1Department of Cardiology, Østfold Hospital, Sarpsborg, Norway; Department of Research, Østfold Hospital, Sarpsborg, Norway.
Journal of thrombosis and haemostasis : JTH
|January 6, 2024
概括
与传统的CAT-BLEED分数相比,机器学习模型显著改善了癌症相关血栓症 (CAT) 患者的出血风险预测. 这一进步为个性化抗凝策略和更好的患者结果提供了潜力.
科学领域:
- 在瘤学瘤学.
- 血液学 血液学 血液学
- 医疗信息学 医疗信息学
背景情况:
- 与癌症相关的血栓形成 (CAT) 是一个重大的临床挑战.
- 目前在CAT中对出血风险的评估仅依赖于CAT-BLEED得分.
- 需要更准确,更动态的风险预测工具.
研究的目的:
- 开发和验证基于机器学习 (ML) 的模型,用于预测CAT患者的出血事件.
- 将ML模型的预测性能与已建立的CAT-BLEED得分进行比较.
主要方法:
- 分析了1080名CAT患者的队列,包括488个临床,生化和诊断属性.
- 使用了机器学习算法,包括Ridge/Lasso逻辑回归,随机森林和极端梯度提升 (XGBoost).
- 模型的性能通过在静脉血栓栓塞 (VTE) 后的不同时间点 (1-90天,1-365天,90-455天) 进行预测出血事件 (重大或临床相关的非重大) 的比较来评估.
主要成果:
- ML模型,特别是拉索逻辑回归和XGBoost,在VTE后1~90天和1~365天内出血事件的CAT-BLEED得分上表现出优异的预测性能.
- 对于1至90天的出血预测,Lasso和XGBoost实现了接收器操作特征曲线 (AUROC) 下的面积为0.64±0.12,相比之下,CAT-BLEED得分为0.48±0.13.
- 对于1-365天的出血预测,Lasso和XGBoost分别实现了0.64±0.08和0.59±0.08的AUROC,超过了CAT-BLEED得分的0.47±0.08.
结论:
- 这项研究引入了第一个基于ML的风险模型,用于预测抗凝血治疗的CAT患者的出血.
- 开发的ML模型在预测准确性方面明显优于传统的CAT-BLEED得分.
- 这种新的方法有望实现个性化的抗凝策略,并改善CAT患者的临床结果.
相关概念视频
Anticoagulant Drugs: Low-Molecular-Weight Heparins
702
Hemostasis is a crucial process that prevents excessive blood loss from damaged blood vessels. It involves various mechanisms such as vasoconstriction, platelet adhesion and activation, and fibrin formation. The importance of each mechanism depends on the type of vessel injury. In contrast, thrombosis is the abnormal formation of a blood clot within the blood vessels, leading to potential complications if the clot obstructs blood flow. Thrombosis can be caused by increased coagulability of the...
702
Disorders of Hemostasis
922
Hemostasis, the process that stops bleeding after a blood vessel injury, is crucial for maintaining the integrity of the circulatory system. However, disorders of hemostasis can disrupt this delicate balance, leading to either excessive clotting or bleeding. These disorders can be broadly classified into thromboembolic disorders and bleeding disorders.
Thromboembolic Disorders
Two factors primarily cause thromboembolic conditions.
Thromboembolic Disorders
Two factors primarily cause thromboembolic conditions.
922


