Development of training environment for deep learning with medical images on supercomputer system based on asynchronous parallel Bayesian optimization
Development of training environment for deep learning with medical images on supercomputer system based on asynchronous parallel Bayesian optimization
复制标题
DOI：
10.1007/s11227-020-03164-7
复制
发表时间：
2020-01
期刊：
影响因子：
--
通讯作者：
Y. Nomura;Issei Sato;T. Hanawa;S. Hanaoka;Takahiro Nakao;T. Takenaga;Tetsuya Hoshino;Yuji Sekiya;S. Miki;T. Yoshikawa;N. Hayashi;O. Abe
中科院分区：
文献类型：
--
作者：
Y. Nomura;Issei Sato;T. Hanawa;S. Hanaoka;Takahiro Nakao;T. Takenaga;Tetsuya Hoshino;Yuji Sekiya;S. Miki;T. Yoshikawa;N. Hayashi;O. Abe
最近，深度学习已经在医学图像分析领域得到了利用。然而，使用医学图像训练深度学习模型非常耗时，因为大多数医学图像数据都是三维体积或高分辨率二维图像。此外，众多超参数的优化强烈影响深度学习的性能。如果能够在超级计算机系统上实现利用超参数优化训练深度学习的框架，有望加速利用医学图像进行深度学习的训练。在这项研究中，我们描述了我们在我们研究所的超级计算机系统（Reedge-H超级计算机系统）上使用医学图像训练深度学习的新环境，该系统基于异步并行贝叶斯优化。我们在构建的环境中训练了两种类型的自动病变检测应用程序。构建的环境使我们能够在短时间内通过超参数调整来训练深度学习。
Recently, deep learning has been exploited in the field of medical image analysis. However, the training of deep learning models with medical images is time-consuming since most medical image data are three-dimensional volumes or high-resolution two-dimensional images. Moreover, the optimization of numerous hyperparameters strongly affects the performance of deep learning. If a framework for training deep learning with hyperparameter optimization on a supercomputer system can be realized, it is expected to accelerate the training of deep learning with medical images. In this study, we described our novel environment for training deep learning with medical images on the supercomputer system in our institute (Reedbush-H supercomputer system) based on asynchronous parallel Bayesian optimization. We trained two types of automated lesion detection application in a constructed environment. The constructed environment enabled us to train deep learning with hyperparameter tuning in a short time.