| Creators: |
Hieber, Daniel and Jungbäck, Nicola and Holl, Felix and Märkl, Bruno and Pryss, Rüdiger and Schobel, Johannes and Liesche-Starnecker, Friederike |
| Title: |
Why We Don’t Save Whole-Slide Images as Lego Mosaics –Putting the Scale of Whole-Slide Images into Perspective |
| Item Type: |
Conference or Workshop Item |
| Event Title: |
(Meeting Abstracts of the) 68th Annual Meeting of the German Society of Neuropathology and Neuroanatomy (DGNN) |
| Event Location: |
Regensburg, Germany |
| Event Dates: |
12.-14. September 2024 |
| Projects: |
DigiHealth, NAP |
| Page Range: |
p. 46 |
| Additional Information: |
Open Access | Free Neuropathology 5:19(2024) p.46 |
| Date: |
2024 |
| Divisions: |
Gesundheitsmanagement |
| Abstract (ENG): |
Why do we save Whole-Slide images (WSIs) as DICOM-WSI, TIFF, SVS, ... and not as Lego mosaics? While the answer to this question might be pretty obvious (working with digital data is much easier, and the staining is closer to the original), we want to use the latter to put the scale of WSIs into perspective.WSIs typically scale up to 80,000 x 60,000 pixels (larger sizes are possible). When transforming such images into Lego, we need 4.8 billion 1 x 1 Lego plates and 4,687,500 32 x 32 Lego base plates. This would accumulate to a total cost of 330 millionEuros for a single WSI using original Lego blocks. The final mosaic would result in asize of 625 m by 468.75 m, a total area of 292,968.75 m2(41.03 soccer fields or approx. 2.09 times Legoland Germany). Since Lego currently offers only 40 different colors for its plates –of which only about 10 to 20 are relevant for pathology (depending on the staining method) –the conversion to a Lego mosaic would result in a significant loss of information.While these numbers may seem absurdly large, processing a digital WSI with all its information at once using a machine learning model would result in 14.4 billion input parameters.While converting a WSI to Lego does not make sense from a practical view point, it illustrates the enormous amount of data in such images. These numbers further demonstrate the immense value of digital pathology, allowing computers to process such data quantities for diagnostic uses.Meeting Abstract |
| Forthcoming: |
No |
| Language: |
English |
| Link eMedia: |
Download |
| Citation: |
Hieber, Daniel and Jungbäck, Nicola and Holl, Felix and Märkl, Bruno and Pryss, Rüdiger and Schobel, Johannes and Liesche-Starnecker, Friederike
(2024)
Why We Don’t Save Whole-Slide Images as Lego Mosaics –Putting the Scale of Whole-Slide Images into Perspective.
In: (Meeting Abstracts of the) 68th Annual Meeting of the German Society of Neuropathology and Neuroanatomy (DGNN), 12.-14. September 2024, Regensburg, Germany, p. 46.
|
 |
View Item in edit mode (academic staff only) |
