Enhanced Diagnostic Fidelity in Pathology Whole Slide Image Compression via Deep Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fischer, Maximilian, Neher, Peter, Schüffler, Peter, Xiao, Shuhan, Almeida, Silvia Dias, Ulrich, Constantin, Muckenhuber, Alexander, Braren, Rickmer, Götz, Michael, Kleesiek, Jens, Nolden, Marco, Maier-Hein, Klaus
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929759895683072
author Fischer, Maximilian
Neher, Peter
Schüffler, Peter
Xiao, Shuhan
Almeida, Silvia Dias
Ulrich, Constantin
Muckenhuber, Alexander
Braren, Rickmer
Götz, Michael
Kleesiek, Jens
Nolden, Marco
Maier-Hein, Klaus
author_facet Fischer, Maximilian
Neher, Peter
Schüffler, Peter
Xiao, Shuhan
Almeida, Silvia Dias
Ulrich, Constantin
Muckenhuber, Alexander
Braren, Rickmer
Götz, Michael
Kleesiek, Jens
Nolden, Marco
Maier-Hein, Klaus
contents Accurate diagnosis of disease often depends on the exhaustive examination of Whole Slide Images (WSI) at microscopic resolution. Efficient handling of these data-intensive images requires lossy compression techniques. This paper investigates the limitations of the widely-used JPEG algorithm, the current clinical standard, and reveals severe image artifacts impacting diagnostic fidelity. To overcome these challenges, we introduce a novel deep-learning (DL)-based compression method tailored for pathology images. By enforcing feature similarity of deep features between the original and compressed images, our approach achieves superior Peak Signal-to-Noise Ratio (PSNR), Multi-Scale Structural Similarity Index (MS-SSIM), and Learned Perceptual Image Patch Similarity (LPIPS) scores compared to JPEG-XL, Webp, and other DL compression methods.
format Preprint
id arxiv_https___arxiv_org_abs_2503_11350
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhanced Diagnostic Fidelity in Pathology Whole Slide Image Compression via Deep Learning
Fischer, Maximilian
Neher, Peter
Schüffler, Peter
Xiao, Shuhan
Almeida, Silvia Dias
Ulrich, Constantin
Muckenhuber, Alexander
Braren, Rickmer
Götz, Michael
Kleesiek, Jens
Nolden, Marco
Maier-Hein, Klaus
Image and Video Processing
Accurate diagnosis of disease often depends on the exhaustive examination of Whole Slide Images (WSI) at microscopic resolution. Efficient handling of these data-intensive images requires lossy compression techniques. This paper investigates the limitations of the widely-used JPEG algorithm, the current clinical standard, and reveals severe image artifacts impacting diagnostic fidelity. To overcome these challenges, we introduce a novel deep-learning (DL)-based compression method tailored for pathology images. By enforcing feature similarity of deep features between the original and compressed images, our approach achieves superior Peak Signal-to-Noise Ratio (PSNR), Multi-Scale Structural Similarity Index (MS-SSIM), and Learned Perceptual Image Patch Similarity (LPIPS) scores compared to JPEG-XL, Webp, and other DL compression methods.
title Enhanced Diagnostic Fidelity in Pathology Whole Slide Image Compression via Deep Learning
topic Image and Video Processing
url https://arxiv.org/abs/2503.11350