Histo-Fetch - On-the-Fly Processing of Gigapixel Whole Slide Images Simplifies and Speeds Neural Network Training.
Histo-Fetch - On-the-Fly Processing of Gigapixel Whole Slide Images Simplifies and Speeds Neural Network Training.
复制标题
DOI：
10.4103/jpi.jpi_59_20
复制
发表时间：
2022
影响因子：
--
通讯作者：
Sarder P
中科院分区：
文献类型：
--
作者：
Lutnick B;Murali LK;Ginley B;Rosenberg AZ;Sarder P
使用病理学全切片图像（WSI）训练卷积神经网络传统上以提取图像块的训练数据集为开端。虽然有效，但对于WSI的大型数据集，这种数据集准备是低效的。我们创建了一个自定义管道（histo-fetch），以有效地从病理WSI中提取随机补丁和标签，以实时输入到神经网络。我们在网络训练过程中根据需要预取这些补丁，避免了WSI准备的需要，如切碎/平铺。我们展示了这个管道的效用，分别使用流行的网络CycleGAN和ProGAN执行人工染色转移和图像生成。对于大型WSI数据集，histo-fetch开始训练的速度快了98.6%，并且使用的磁盘空间减少了7535倍。
Training convolutional neural networks using pathology whole slide images (WSIs) is traditionally prefaced by the extraction of a training dataset of image patches. While effective, for large datasets of WSIs, this dataset preparation is inefficient. We created a custom pipeline (histo-fetch) to efficiently extract random patches and labels from pathology WSIs for input to a neural network on-the-fly. We prefetch these patches as needed during network training, avoiding the need for WSI preparation such as chopping/tiling. We demonstrate the utility of this pipeline to perform artificial stain transfer and image generation using the popular networks CycleGAN and ProGAN, respectively. For a large WSI dataset, histo-fetch is 98.6% faster to start training and used 7535x less disk space.