一种机器学习方法来预测成功的跨腹腔外微侵袭性门修复
Alessandro Vairo1,2, Caterina Russo1, Andrea Saglietto1,2
1Division of Cardiology, Cardiovascular and Thoracic Department, Città della Salute e della Scienza Hospital, 10126 Turin, Italy.
Journal of clinical medicine
|August 28, 2025
概括
一个机器学习模型可以预测NeoChord手术的成功. 该系统使用手术前的临床和心声学数据来识别可能获得良好结果的患者.
科学领域:
- 心血管外科
- 医学成像
- 医学中的机器学习
背景情况:
- NeoChord手术是一种心跳技术,用于退行性腹 (MR).
- 它使用跨腹腔的方法来治疗由叶片落或落引起的MR.
- 声心脏成像指导了这一过程.
研究的目的:
- 开发一个机器学习 (ML) 评分系统来预测NeoChord程序的成功.
- 成功被定义为在随访时达到不到中度的MR.
- 该系统使用了手术前的临床和心声数据.
主要方法:
- 分析了接受NeoChord手术的80名患者.
- 通过三维食管回声扫描评估了手术前的 mitra 门解剖学.
- 监督的ML模型 (随机森林,决策树) 被训练并验证.
主要成果:
- 随机森林模型实现了0.79 (平均值) 和0.83 (中位数) 的曲线下的交叉验证面积.
- 决策树模型实现了0.72 (平均值) 和0.77 (中位数) 的曲线下的交叉验证面积.
- 关键预测因素包括年龄,肌间隙,早期心肌区域和左心房体积指数.
结论:
- 机器学习可以有效地预测NeoChord技术的中期成功.
- 整合临床和3D心声学数据可以提高预测的准确性.
- 这种方法可能有助于在手术前为NeoChord手术选择患者.
相关概念视频
Mitral Stenosis III: Medical Management
21
Mitral stenosis, a condition marked by the narrowing of the mitral valve, necessitates an integrated approach for effective management. This approach includes preventative measures, medical therapy, and surgical interventions to reduce symptoms and prevent complications.PreventionPrevention of mitral stenosis primarily focuses on reducing the incidence of bacterial infections, particularly streptococcal infections, which can lead to rheumatic fever and subsequent valvular damage. Timely...
21
Mitral Valve Prolapse II: Assessment and Management
39
IntroductionA range of clinical features characterizes Mitral Valve Prolapse (MVP), but it is important to note that many individuals with MVP are asymptomatic and may remain so throughout their lives. For those who do exhibit symptoms, the following are the key clinical features:Palpitations: This is a common symptom where individuals feel an irregular or rapid heartbeat. Palpitations in MVP are often due to arrhythmias such as premature ventricular contractions or supraventricular...
39


