Mehta Vishal S, Ma YingLiang, Wijesuriya Nadeev, DeVere Felicity, Howell Sandra, Elliott Mark K, Mannkakara Nilanka N, Hamakarim Tatiana, Wong Tom, O'Brien Hugh, Niederer Steven, Razavi Reza, Rinaldi Christopher A
Cardiology Department, Guy's and St Thomas' NHS Foundation Trust, London, United Kingdom; School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom.
School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom; School of Computing Sciences, University of East Anglia, Norwich, United Kingdom.
Heart Rhythm. 2024 Jun;21(6):919-928. doi: 10.1016/j.hrthm.2024.02.015. Epub 2024 Feb 12.
Machine learning (ML) models have been proposed to predict risk related to transvenous lead extraction (TLE).
The purpose of this study was to test whether integrating imaging data into an existing ML model increases its ability to predict major adverse events (MAEs; procedure-related major complications and procedure-related deaths) and lengthy procedures (≥100 minutes).
We hypothesized certain features-(1) lead angulation, (2) coil percentage inside the superior vena cava (SVC), and (3) number of overlapping leads in the SVC-detected from a pre-TLE plain anteroposterior chest radiograph (CXR) would improve prediction of MAE and long procedural times. A deep-learning convolutional neural network was developed to automatically detect these CXR features.
A total of 1050 cases were included, with 24 MAEs (2.3%) . The neural network was able to detect (1) heart border with 100% accuracy; (2) coils with 98% accuracy; and (3) acute angle in the right ventricle and SVC with 91% and 70% accuracy, respectively. The following features significantly improved MAE prediction: (1) ≥50% coil within the SVC; (2) ≥2 overlapping leads in the SVC; and (3) acute lead angulation. Balanced accuracy (0.74-0.87), sensitivity (68%-83%), specificity (72%-91%), and area under the curve (AUC) (0.767-0.962) all improved with imaging biomarkers. Prediction of lengthy procedures also improved: balanced accuracy (0.76-0.86), sensitivity (75%-85%), specificity (63%-87%), and AUC (0.684-0.913).
Risk prediction tools integrating imaging biomarkers significantly increases the ability of ML models to predict risk of MAE and long procedural time related to TLE.
已提出机器学习（ML）模型来预测与经静脉导线拔除术（TLE）相关的风险。
本研究的目的是测试将影像数据整合到现有的ML模型中是否能提高其预测主要不良事件（MAE；与手术相关的主要并发症和与手术相关的死亡）和冗长手术（≥100分钟）的能力。
我们假设从TLE前的胸部正位X线片（CXR）中检测到的某些特征——（1）导线成角、（2）上腔静脉（SVC）内的线圈百分比、（3）SVC内重叠导线的数量——将改善对MAE和长时间手术的预测。开发了一种深度学习卷积神经网络来自动检测这些CXR特征。
共纳入1050例病例，其中24例发生MAE（2.3%）。神经网络能够分别以100%的准确率检测心脏边界、以98%的准确率检测线圈、以91%和70%的准确率检测右心室和SVC中的锐角。以下特征显著改善了MAE预测：（1）SVC内线圈≥50%；（2）SVC内≥2条重叠导线；（3）导线锐角成角。影像生物标志物使平衡准确率（0.74 - 0.87）、敏感性（68% - 83%）、特异性（72% - 91%）和曲线下面积（AUC）（0.767 - 0.962）均得到改善。对冗长手术的预测也得到改善：平衡准确率（0.76 - 0.86）、敏感性（75% - 85%）、特异性（63% - 87%）和AUC（0.684 - 0.913）。
整合影像生物标志物的风险预测工具显著提高了ML模型预测与TLE相关的MAE风险和长时间手术的能力。
