skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Scaling Resolution of Gigapixel Whole Slide Images Using Spatial Decomposition on Convolutional Neural Networks

Conference ·
 [1]; ORCiD logo [1];  [1]; ORCiD logo [1]; ORCiD logo [1]; ORCiD logo [1];  [2]
  1. ORNL
  2. NVIDIA, Santa Clara, CA
+ Show Author Affiliations

Gigapixel images are prevalent in scientific domains ranging from remote sensing, and satellite imagery to microscopy, etc. However, training a deep learning model at the natural resolution of those images has been a challenge in terms of both, overcoming the resource limit (e.g. HBM memory constraints), as well as scaling up to a large number of GPUs. In this paper, we trained Residual neural Networks (ResNet) on 22,528 x 22,528-pixel size images using a distributed spatial decomposition method on 2,304 GPUs on the Summit Supercomputer. We applied our method on a Whole Slide Imaging (WSI) dataset from The Cancer Genome Atlas (TCGA) database. WSI images can be in the size of 100,000 x 100,000 pixels or even larger, and in this work we studied the effect of image resolution on a classification task, while achieving state-of-the-art AUC scores. Moreover, our approach doesn't need pixel-level labels, since we're avoiding patching from the WSI images completely, while adding the capability of training arbitrary large-size images. This is achieved through a distributed spatial decomposition method, by leveraging the non-block fat-tree interconnect network of the Summit architecture, which enabled GPU-to-GPU direct communication. Finally, detailed performance analysis results are shown, as well as a comparison with a data-parallel approach when possible.

Research Organization:
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Sponsoring Organization:
USDOE
DOE Contract Number:
AC05-00OR22725
OSTI ID:
1997785
Resource Relation:
Conference: The Platform for Advanced Scientific Computing (PASC) - Davos, , Switzerland - 6/26/2023 12:00:00 PM-6/28/2023 12:00:00 PM
Country of Publication:
United States
Language:
English

References (11)

ImageNet: A large-scale hierarchical image database
  • Deng, Jia; Dong, Wei; Socher, Richard
  • 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2009 IEEE Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.2009.5206848
conference June 2009
Improving Strong-Scaling of CNN Training by Exploiting Finer-Grained Parallelism conference May 2019
Toward Large-Scale Image Segmentation on Summit
  • Seal, Sudip K.; Lim, Seung-Hwan; Wang, Dali
  • ICPP '20: 49th International Conference on Parallel Processing, 49th International Conference on Parallel Processing - ICPP https://doi.org/10.1145/3404397.3404468
conference August 2020
Distributed Training for High Resolution Images: A Domain and Spatial Decomposition Approach conference November 2021
An annotation-free whole-slide training approach to pathological classification of lung cancer types using deep learning journal February 2021
Data-efficient and weakly supervised computational pathology on whole-slide images journal March 2021
Multiple instance learning of deep convolutional neural networks for breast histopathology whole slide classification conference April 2018
Deep Multi-instance Learning for Survival Prediction from Whole Slide Images book January 2019
Scaling Vision Transformers to Gigapixel Images via Hierarchical Self-Supervised Learning conference June 2022
Streaming Convolutional Neural Networks for End-to-End Learning With Multi-Megapixel Images journal March 2022
Scaling the Summit: Deploying the World’s Fastest Supercomputer book January 2019

Similar Records

Related Subjects