Deep Learning on Lossily Compressed Pathology Images: Adverse Effects for ImageNet Pre-trained Models

Maximilian Fischer, Peter Neher, Michael Götz, Shuhan Xiao, Silvia Dias Almeida, Peter Schüffler, Alexander Muckenhuber, Rickmer Braren, Jens Kleesiek, Marco Nolden, Klaus Maier-Hein

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Digital Whole Slide Imaging (WSI) systems allow scanning complete probes at microscopic resolutions, making image compression inevitable to reduce storage costs. While lossy image compression is readily incorporated in proprietary file formats as well as the open DICOM format for WSI, its impact on deep-learning algorithms is largely unknown. We compare the performance of several deep learning classification architectures on different datasets using a wide range and different combinations of compression ratios during training and inference. We use ImageNet pre-trained models, which is commonly applied in computational pathology. With this work, we present a quantitative assessment on the effects of repeated lossy JPEG compression for ImageNet pre-trained models. We show adverse effects for a classification task, when certain quality factors are combined during training and inference.

Original languageEnglish
Title of host publicationMedical Optical Imaging and Virtual Microscopy Image Analysis - 1st International Workshop, MOVI 2022, Held in Conjunction with MICCAI 2022, Proceedings
EditorsYuankai Huo, Bryan A. Millis, Yuyin Zhou, Xiangxue Wang, Adam P. Harrison, Ziyue Xu
PublisherSpringer Science and Business Media Deutschland GmbH
Pages73-83
Number of pages11
ISBN (Print)9783031169601
DOIs
StatePublished - 2022
Externally publishedYes
Event1st International Workshop on Medical Optical Imaging and Virtual Microscopy Image Analysis, MOVI 2022, held in conjunction with the 25th International Conference on Medical Imaging and Computer Assisted Intervention, MICCAI 2022 - Singapore, Singapore
Duration: 18 Sep 202218 Sep 2022

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13578 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference1st International Workshop on Medical Optical Imaging and Virtual Microscopy Image Analysis, MOVI 2022, held in conjunction with the 25th International Conference on Medical Imaging and Computer Assisted Intervention, MICCAI 2022
Country/TerritorySingapore
CitySingapore
Period18/09/2218/09/22

Keywords

  • Compression artifacts
  • Pathology image classification
  • Whole Slide Imaging

Fingerprint

Dive into the research topics of 'Deep Learning on Lossily Compressed Pathology Images: Adverse Effects for ImageNet Pre-trained Models'. Together they form a unique fingerprint.

Cite this