TY - GEN
T1 - Deep Learning on Lossily Compressed Pathology Images
T2 - 1st 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
AU - Fischer, Maximilian
AU - Neher, Peter
AU - Götz, Michael
AU - Xiao, Shuhan
AU - Almeida, Silvia Dias
AU - Schüffler, Peter
AU - Muckenhuber, Alexander
AU - Braren, Rickmer
AU - Kleesiek, Jens
AU - Nolden, Marco
AU - Maier-Hein, Klaus
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - 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.
AB - 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.
KW - Compression artifacts
KW - Pathology image classification
KW - Whole Slide Imaging
UR - http://www.scopus.com/inward/record.url?scp=85138784528&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-16961-8_8
DO - 10.1007/978-3-031-16961-8_8
M3 - Conference contribution
AN - SCOPUS:85138784528
SN - 9783031169601
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 73
EP - 83
BT - Medical Optical Imaging and Virtual Microscopy Image Analysis - 1st International Workshop, MOVI 2022, Held in Conjunction with MICCAI 2022, Proceedings
A2 - Huo, Yuankai
A2 - Millis, Bryan A.
A2 - Zhou, Yuyin
A2 - Wang, Xiangxue
A2 - Harrison, Adam P.
A2 - Xu, Ziyue
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 18 September 2022 through 18 September 2022
ER -