Lossless Compression of JPEG2000 Whole Slide Images Is Not Required for Diagnostic Virtual Microscopy

Abstract

The use of lossy compression in medical imaging is controversial, although it is inevitable to reduce large data amounts. In contrast with lossy compression, lossless compression does not impair image quality. In addition to our previous studies, we evaluated virtual 3-dimensional microscopy using JPEG2000 whole slide images of gastric biopsy specimens with or without Helicobacter pylori gastritis using lossless compression (1:1) or lossy compression with different compression levels: 5:1, 10:1, and 20:1. The virtual slides were diagnosed in a blinded manner by 3 pathologists using the updated Sydney classification. The results showed no significant differences in the diagnosis of H pylori between the different levels of compression in virtual microscopy. We assume that lossless compression is not required for diagnostic virtual microscopy. The limits of lossy compression in virtual microscopy without a loss of diagnostic quality still need to be determined. Analogous to the processes in radiology, recommendations for the use of lossy compression in diagnostic virtual microscopy have to be worked out by pathology societies.

Impact of JPEG 2000 compression on deep convolutional neural networks for metastatic cancer detection in histopathological images

“…Luckily, it is generally not prohibited by the main regulatory bodies in the European Union, United States, Canada, and Australia, provided that it does not impair the diagnostic quality and does not cause new risks compared with conventional practice. 8 Hence, it is important to define a strategy or protocol for an efficient parameterization of the deployed compression techniques to yield a high compression ratio without jeopardizing the classification performance. The issue of higher compression ratio with lower encoding time has been recognized as well in recent efforts for creating the DICOM standard in the field of digital pathology.…”

“…9 For studying the impact of lossy compression on the diagnostic performance of human experts, several studies have been reported. 8,[10][11][12][13][14][15][16] Mostly, they reported that the human visual perception is to some extent robust against image quality degradation. However, there is not a generally accepted tolerance level with respect to the diagnostic accuracy.…”

“…In addition, because clinical evaluations can be subjective and have a bias regarding the task at hand and the skill of experts, different studies have suggested different compression ratios corresponding to the addressed clinical task. For example, Kalinski et al 8 reported that the impact of a JPEG 2000 compression ratio up to 20 did not show significant influence on the detection of Helicobacter pylori gastritis in gastric histopathological images, performed by three pathologists. In another work by Krupinski et al, 11 by involving six pathologists, a compression factor of up to 32 did not cause noticeable difference in distinguishing benign from malignant cancer in breast tissue.…”

SlideToolkit: An Assistive Toolset for the Histological Quantification of Whole Slide Images

“…Commonly used lossy image compression is JPEG and an example of a lossless compression is PNG. High-quality lossy compression can be sufficient for image processing, [14] but lossless compression is recommended. [8] We used 90% lossy compression (JPEG format) for the 20x WSIs which was sufficient for our analysis; lower quality settings were not tested.…”

How to Prepare Neuroanatomical Image Data

“…For large virtual tissues, a format that satisfies many limited bandwidth and image fidelity requirements is the JPEG2000 standard, where compression ratios of thirty-to-one can often be used without a large sacrifice in image quality (Tuominen and Isola, 2009). This standard has also been successfully applied in digital pathology using a twenty-to-one compression ratio of three-dimensional whole-slide pathology images (Kalinski et al, 2011). Current formats that support the JPEG2000 standard are MBF Bioscience's JPX and Aperio's SVS formats.…”