A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions.
A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions.
复制标题
DOI：
10.1038/s41467-021-24483-0
复制
发表时间：
2021-07-09
影响因子：
16.6
通讯作者：
Clifton D
中科院分区：
文献类型：
--
作者：
Kiyasseh D;Zhu T;Clifton D
众所周知，在违反独立同分布 (i.i.d.) 假设的实例上训练的深度学习算法会经历破坏性干扰，这种现象的特征是性能下降。然而，这种违规行为在临床环境中普遍存在，其中数据从不同的临床站点和多个生理传感器暂时传输。为了减轻这种干扰，我们提出了一种名为 CLOPS 的持续学习策略，该策略采用重播缓冲区。为了引导实例存储到缓冲区中，我们提出了端到端可训练参数，称为任务实例参数，它量化了深度学习系统对数据点进行分类的难度。我们通过临床领域知识验证这些参数的解释。为了重放缓冲区中的实例，我们利用基于不确定性的采集函数。在反映疾病、时间、数据模式和医疗机构转变的四个持续学习场景中的三个场景中，我们表明 CLOPS 优于最先进的方法、GEM 和 MIR。我们还进行了广泛的消融研究，以证明我们提出的策略各个组成部分的必要性。我们的框架有可能为长期保持稳健的诊断系统铺平道路。根据来自不同临床站点和大量生理传感器的临时流数据进行训练的深度学习算法通常会受到性能下降的影响。为了缓解这个问题，作者提出了一种采用重放缓冲区的持续学习策略。
Deep learning algorithms trained on instances that violate the assumption of being independent and identically distributed (i.i.d.) are known to experience destructive interference, a phenomenon characterized by a degradation in performance. Such a violation, however, is ubiquitous in clinical settings where data are streamed temporally from different clinical sites and from a multitude of physiological sensors. To mitigate this interference, we propose a continual learning strategy, entitled CLOPS, that employs a replay buffer. To guide the storage of instances into the buffer, we propose end-to-end trainable parameters, termed task-instance parameters, that quantify the difficulty with which data points are classified by a deep-learning system. We validate the interpretation of these parameters via clinical domain knowledge. To replay instances from the buffer, we exploit uncertainty-based acquisition functions. In three of the four continual learning scenarios, reflecting transitions across diseases, time, data modalities, and healthcare institutions, we show that CLOPS outperforms the state-of-the-art methods, GEM and MIR. We also conduct extensive ablation studies to demonstrate the necessity of the various components of our proposed strategy. Our framework has the potential to pave the way for diagnostic systems that remain robust over time. Deep learning algorithms trained on data streamed temporally from different clinical sites and from a multitude of physiological sensors are generally affected by a degradation in performance. To mitigate this, the authors propose a continual learning strategy that employs a replay buffer.
登录
查看更多内容
影响因子：
82.9
作者：
Hannun, Awni Y.;Rajpurkar, Pranav;Ng, Andrew Y.
通讯作者：
Ng, Andrew Y.
DOI：
10.22489/cinc.2017.065-469
发表时间：
2017-09
期刊：
Computing in cardiology
影响因子：
--
作者：
Clifford GD;Liu C;Moody B;Lehman LH;Silva I;Li Q;Johnson AE;Mark RG
通讯作者：
Mark RG
影响因子：
9.8
作者：
Martinez-Sancho, Elisabet;Slamova, Lenka;Fonti, Patrick
通讯作者：
Fonti, Patrick
影响因子：
--
作者：
Dwork, Cynthia;Roth, Aaron
通讯作者：
Roth, Aaron