Federated learning and differential privacy for medical image analysis.
Federated learning and differential privacy for medical image analysis.
复制标题
DOI：
10.1038/s41598-022-05539-7
复制
发表时间：
2022-02-04
影响因子：
4.6
通讯作者：
Tizhoosh HR
中科院分区：
文献类型：
--
作者：
Adnan M;Kalra S;Cresswell JC;Taylor GW;Tizhoosh HR
大规模数据集的出现推动了人工智能革命的向前发展。相比之下，大规模医学数据集的缺乏阻碍了机器学习在医疗保健中的应用。缺乏公开可用的、多中心的和多样化的数据集，主要是因为共享医疗数据的机密性和隐私问题。为了展示医学图像成像的一条可行的前进道路，我们进行了一个应用差异私有联合学习框架来分析组织病理学图像的案例研究，组织病理学图像是最大也可能是最复杂的医学图像。我们使用公共存储库癌症基因组图谱(TCGA)数据集来模拟分布式环境，研究了IID和非IID分布以及医疗保健提供者的数量(即医院和诊所)以及个人数据集大小的影响。我们通过实验比较了私密、分布式训练和传统训练的性能，并证明了分布式训练可以在具有较强隐私保证的情况下获得相似的性能。我们还研究了不同源域对组织病理学图像的影响，通过外部验证来评估其性能。我们的工作表明，对于医学图像分析中的机器学习模型的协作开发，差分私有联合学习是一种可行且可靠的框架。
The artificial intelligence revolution has been spurred forward by the availability of large-scale datasets. In contrast, the paucity of large-scale medical datasets hinders the application of machine learning in healthcare. The lack of publicly available multi-centric and diverse datasets mainly stems from confidentiality and privacy concerns around sharing medical data. To demonstrate a feasible path forward in medical image imaging, we conduct a case study of applying a differentially private federated learning framework for analysis of histopathology images, the largest and perhaps most complex medical images. We study the effects of IID and non-IID distributions along with the number of healthcare providers, i.e., hospitals and clinics, and the individual dataset sizes, using The Cancer Genome Atlas (TCGA) dataset, a public repository, to simulate a distributed environment. We empirically compare the performance of private, distributed training to conventional training and demonstrate that distributed training can achieve similar performance with strong privacy guarantees. We also study the effect of different source domains for histopathology images by evaluating the performance using external validation. Our work indicates that differentially private federated learning is a viable and reliable framework for the collaborative development of machine learning models in medical image analysis.
登录
查看更多内容
影响因子：
15.2
作者：
Aggarwal R;Sounderajah V;Martin G;Ting DSW;Karthikesalingam A;King D;Ashrafian H;Darzi A
通讯作者：
Darzi A
DOI：
10.1093/jamia/ocy017
发表时间：
2018-08-01
期刊：
Journal of the American Medical Informatics Association : JAMIA
影响因子：
--
作者：
Chang K;Balachandar N;Lam C;Yi D;Brown J;Beers A;Rosen B;Rubin DL;Kalpathy-Cramer J
通讯作者：
Kalpathy-Cramer J
影响因子：
8
作者：
Carbonneau, Marc-Andre;Cheplygina, Veronika;Gagnon, Ghyslain
通讯作者：
Gagnon, Ghyslain
影响因子：
10.9
作者：
Kalra, Shivam;Tizhoosh, H. R.;Pantanowitz, Liron
通讯作者：
Pantanowitz, Liron
影响因子：
10.9
作者：
Li X;Gu Y;Dvornek N;Staib LH;Ventola P;Duncan JS
通讯作者：
Duncan JS