Weakly supervised classification of aortic valve malformations using unlabeled cardiac MRI sequences
Weakly supervised classification of aortic valve malformations using unlabeled cardiac MRI sequences
复制标题
DOI：
10.1038/s41467-019-11012-3
复制
发表时间：
2019-07-15
影响因子：
16.6
通讯作者：
Priest, James R.
中科院分区：
文献类型：
--
作者：
Fries, Jason A.;Varma, Paroma;Priest, James R.
英国生物银行等生物医学存储库提供了越来越多的前瞻性收集的心脏成像数据，但这些数据未标记，这对其在监督机器学习中的使用造成了障碍。我们开发了一种弱监督深度学习模型，使用多达 4,000 个未标记的心脏 MRI 序列对主动脉瓣畸形进行分类。弱监督不需要精心策划的训练数据，而是依赖于领域专家定义的噪声启发法来以编程方式生成大规模的、不完美的训练标签。对于主动脉瓣分类，使用不完美标签训练的模型明显优于使用手工标记 MRI 训练的监督模型。在使用健康结果数据的正交验证实验中，我们的模型识别出主要不良心脏事件风险增加 1.8 岁的个体。这项工作正式确定了主动脉瓣分类的深度学习基线，并概述了使用弱监督来大规模使用未标记的医学图像训练机器学习模型的一般策略。
Biomedical repositories such as the UK Biobank provide increasing access to prospectively collected cardiac imaging, however these data are unlabeled, which creates barriers to their use in supervised machine learning. We develop a weakly supervised deep learning model for classification of aortic valve malformations using up to 4,000 unlabeled cardiac MRI sequences. Instead of requiring highly curated training data, weak supervision relies on noisy heuristics defined by domain experts to programmatically generate large-scale, imperfect training labels. For aortic valve classification, models trained with imperfect labels substantially outperform a supervised model trained on hand-labeled MRIs. In an orthogonal validation experiment using health outcomes data, our model identifies individuals with a 1.8-old increase in risk of a major adverse cardiac event. This work formalizes a deep learning baseline for aortic valve classification and outlines a general strategy for using weak supervision to train machine learning models using unlabeled medical images at scale.