Distributed deep learning networks among institutions for medical imaging.
Distributed deep learning networks among institutions for medical imaging.
复制标题
DOI：
10.1093/jamia/ocy017
复制
发表时间：
2018-08-01
期刊：
影响因子：
--
通讯作者：
Kalpathy-Cramer J
中科院分区：
文献类型：
--
作者：
Chang K;Balachandar N;Lam C;Yi D;Brown J;Beers A;Rosen B;Rubin DL;Kalpathy-Cramer J
深度学习已经成为自动支持临床诊断的一种很有前途的方法。当医疗数据样本有限时，需要多个机构之间的协作才能实现高算法性能。然而，由于技术、法律的或伦理问题，共享患者数据通常具有限制。在这项研究中，我们提出了分发深度学习模型的方法，作为共享患者数据的一种有吸引力的替代方案。我们使用各种训练算法模拟了4个机构的深度学习模型的分布，并将结果与在集中托管的患者数据上训练的深度学习模型进行了比较。研究的训练策略包括集成单机构模型、单一权重转移和周期性权重转移。我们在3个独立的图像集合（视网膜眼底照片，乳房X光检查和ImageNet）中评估了这些图像分类方法。我们发现，周期性权重转移导致的性能与集中托管的患者数据相当。我们还发现，有一个改进的性能的循环权重转移启发式的权重转移频率高。我们表明，分发深度学习模型是共享患者数据的有效替代方案。这一发现对任何合作式深度学习研究都有意义。
Deep learning has become a promising approach for automated support for clinical diagnosis. When medical data samples are limited, collaboration among multiple institutions is necessary to achieve high algorithm performance. However, sharing patient data often has limitations due to technical, legal, or ethical concerns. In this study, we propose methods of distributing deep learning models as an attractive alternative to sharing patient data. We simulate the distribution of deep learning models across 4 institutions using various training heuristics and compare the results with a deep learning model trained on centrally hosted patient data. The training heuristics investigated include ensembling single institution models, single weight transfer, and cyclical weight transfer. We evaluated these approaches for image classification in 3 independent image collections (retinal fundus photos, mammography, and ImageNet). We find that cyclical weight transfer resulted in a performance that was comparable to that of centrally hosted patient data. We also found that there is an improvement in the performance of cyclical weight transfer heuristic with a high frequency of weight transfer. We show that distributing deep learning models is an effective alternative to sharing patient data. This finding has implications for any collaborative deep learning study.
登录
查看更多内容
影响因子：
3.5
作者：
Samala RK;Chan HP;Hadjiiski LM;Helvie MA;Cha KH;Richter CD
通讯作者：
Richter CD
DOI：
10.1093/jamia/ocx101
发表时间：
2018-01-01
影响因子：
6.4
作者：
Xia, Weiyi;Wan, Zhiyu;Malin, Bradley A.
通讯作者：
Malin, Bradley A.
影响因子：
19.5
作者：
Russakovsky, Olga;Deng, Jia;Fei-Fei, Li
通讯作者：
Fei-Fei, Li
影响因子：
64.8
作者：
Esteva A;Kuprel B;Novoa RA;Ko J;Swetter SM;Blau HM;Thrun S
通讯作者：
Thrun S
影响因子：
4.6
作者：
Miotto R;Li L;Kidd BA;Dudley JT
通讯作者：
Dudley JT