Related Experiment Video
Updated: Jun 6, 2026
06:25
Using Computer Vision Libraries to Streamline Nuclei Quantification
Published on: June 6, 2025
JPEG2000 for automated quantification of immunohistochemically stained cell nuclei: a comparative study with standard
Marylène Lejeune1, Carlos López, Ramón Bosch
1Molecular Biology and Research Section, Hospital de Tortosa Verge de la Cinta, IISPV, URV, c/Esplanetes 14, 43500, Tortosa, Spain. mlejeune.htvc.ics@gmail.com
Virchows Archiv : an International Journal of Pathology
|November 19, 2010
Summary
JPEG2000 compression maintains immunohistochemical analytical quality, offering efficient storage for digital pathology images. This format ensures reliable automated evaluation, even with high complexity and varying compression rates.
Area of Science:
- Digital Pathology
- Image Compression Technologies
- Computational Biology
Background:
- Joint Photographic Experts Group (JPEG) and JPEG2000 are widely used image compression formats.
- JPEG2000 offers improved quality and compression ratios compared to standard JPEG.
- Automated evaluation of immunohistochemically stained images requires assessing compression impact on analytical quality.
Purpose of the Study:
- To compare the variability in automated evaluation of immunohistochemically stained digital images between JPEG and JPEG2000 compression formats.
- To determine the influence of compression rates and image complexity on the accuracy of nuclear marker counts (Ki67, FOXP3).
- To assess the suitability of JPEG2000 for efficient storage without compromising immunohistochemical analytical quality.
Main Methods:
- 329 digital images (Ki67 and FOXP3 markers) were analyzed: uncompressed (TIFF), JPEG compressed (1:3, 1:23, 1:46), and JPEG2000 compressed (1:3, 1:23, 1:46).
- Count differences between TIFF and compressed formats were compared at varying compression levels.
- Impact of image complexity (low vs. high) on count accuracy was evaluated.
Main Results:
- JPEG2000 compression resulted in stained nuclei counts closely matching uncompressed TIFF images, particularly for complex images at lower compression ratios.
- For low-complexity images, both JPEG and JPEG2000 showed similar count efficiency to TIFF across all compression levels.
- JPEG2000 demonstrated superior performance in maintaining analytical quality compared to JPEG at higher compression ratios.
Conclusions:
- JPEG2000 compression is a viable option for digital pathology image storage, offering significant file size reduction without compromising immunohistochemical analytical quality.
- The choice of compression format impacts automated image analysis, with JPEG2000 being more robust than JPEG.
- JPEG2000 facilitates efficient data management in digital pathology workflows, ensuring reliable quantitative analysis.
