Towards a privacy-preserving distributed cloud service for preprocessing very large medical images

TL;DR

This paper presents a cloud-based, privacy-preserving distributed preprocessing service for large medical images, enhancing efficiency and security in histopathology analysis.

Contribution

It introduces a novel cloud service that parallelizes preprocessing of gigapixel medical images while ensuring privacy through random tile distribution.

Findings

Improved processing efficiency for large medical images.

Enhanced privacy preservation during distributed processing.

Integration with Jupyter VRE for user-friendly automation.

Abstract

Digitized histopathology glass slides, known as Whole Slide Images (WSIs), are often several gigapixels large and contain sensitive metadata information, which makes distributed processing unfeasible. Moreover, artifacts in WSIs may result in unreliable predictions when directly applied by Deep Learning (DL) algorithms. Therefore, preprocessing WSIs is beneficial, e.g., eliminating privacy-sensitive information, splitting a gigapixel medical image into tiles, and removing the diagnostically irrelevant areas. This work proposes a cloud service to parallelize the preprocessing pipeline for large medical images. The data and model parallelization will not only boost the end-to-end processing efficiency for histological tasks but also secure the reconstruction of WSI by randomly distributing tiles across processing nodes. Furthermore, the initial steps of the pipeline will be integrated into the Jupyter-based Virtual Research Environment (VRE) to enable image owners to configure and automate the execution process based on resource allocation.