Decoodt Pierre, Sierra-Sosa Daniel, Anghel Laura, Cuminetti Giovanni, De Keyzer Eva, Morissens Marielle
Cardiologie, Centre Hospitalier Universitaire Brugmann, Faculté de Médecine, Université Libre de Bruxelles, Place Van Gehuchten 4, 1020 Brussels, Belgium.
Computer Science and Information Technologies Department, Hood College, 401 Rosemont Ave., Frederick, MD 21702, USA.
Diagnostics (Basel). 2024 Jul 5;14(13):1439. doi: 10.3390/diagnostics14131439.
Identifying patients with left ventricular ejection fraction (EF), either reduced [EF < 40% (rEF)], mid-range [EF 40-50% (mEF)], or preserved [EF > 50% (pEF)], is considered of primary clinical importance. An end-to-end video classification using AutoML in Google Vertex AI was applied to echocardiographic recordings. Datasets balanced by majority undersampling, each corresponding to one out of three possible classifications, were obtained from the Standford EchoNet-Dynamic repository. A train-test split of 75/25 was applied. A binary video classification of rEF vs. not rEF demonstrated good performance (test dataset: ROC AUC score 0.939, accuracy 0.863, sensitivity 0.894, specificity 0.831, positive predicting value 0.842). A second binary classification of not pEF vs. pEF was slightly less performing (test dataset: ROC AUC score 0.917, accuracy 0.829, sensitivity 0.761, specificity 0.891, positive predicting value 0.888). A ternary classification was also explored, and lower performance was observed, mainly for the mEF class. A non-AutoML PyTorch implementation in open access confirmed the feasibility of our approach. With this proof of concept, end-to-end video classification based on transfer learning to categorize EF merits consideration for further evaluation in prospective clinical studies.
识别左心室射血分数（EF）降低[EF<40%（rEF）]、中等范围[EF 40-50%（mEF）]或保留[EF>50%（pEF）]的患者被认为具有重要的临床意义。在谷歌Vertex AI中使用自动机器学习进行的端到端视频分类被应用于超声心动图记录。通过多数欠采样平衡的数据集，每个数据集对应三种可能分类中的一种，这些数据集来自斯坦福EchoNet-Dynamic存储库。采用75/25的训练-测试分割。rEF与非rEF的二元视频分类表现良好（测试数据集：ROC AUC分数0.939，准确率0.863，灵敏度0.894，特异性0.831，阳性预测值0.842）。非pEF与pEF的第二个二元分类表现略差（测试数据集：ROC AUC分数0.917，准确率0.829，灵敏度0.761，特异性0.891，阳性预测值0.888）。还探索了三元分类，观察到性能较低，主要是对于mEF类别。开放获取中的非自动机器学习PyTorch实现证实了我们方法的可行性。有了这个概念验证，基于迁移学习对EF进行分类的端到端视频分类值得在前瞻性临床研究中进一步评估。
