Datasets:

Modalities:
Image
Text
Formats:
webdataset
ArXiv:
Libraries:
Datasets
WebDataset
License:
lely475 commited on
Commit
b1ee185
·
2 Parent(s): 78e792e e5bf357
Files changed (1) hide show
  1. README.md +13 -14
README.md CHANGED
@@ -1,7 +1,5 @@
1
  ---
2
  license: cc-by-4.0
3
- task_categories:
4
- - feature-extraction
5
  pretty_name: >-
6
  Revisiting Automatic Data Curation for Vision Foundation Models in Digital
7
  Pathology
@@ -13,20 +11,21 @@ pretty_name: >-
13
 
14
  ## Data
15
  We provide the following data:
16
- 1. **Clustering result:** Hierarchical cluster labels for each of the 350M tiles for mode t1 and t2 as csv files. The csv contains the slide_id (WSI the tile originates from), (x,y)-coordinate of the tile at level 0 (highest pyramid level of the WSI) and cluster labels for each level (denoted as columns "level_1"..."level_4"). Then slides can be downloaded from the [TCGA](https://portal.gdc.cancer.gov/) and [GTEx](https://www.gtexportal.org/home/histologyPage) websites, for tile extraction from the WSIs we recommend [openslide](https://openslide.org/api/python/).
17
- Structure of `clustering_results/clustering_{t1,t2}.csv`:
 
18
 
19
- | slide\_id | tile\_x | tile\_y | level\_1 | level\_2 | level\_3 | level\_4 |
20
- | ------------------------------------------------- | ------- | ------- | -------- | -------- | -------- | -------- |
21
- | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 32406 | 10621 | 1301309 | 17404 | 2 | 24 |
22
- | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 32850 | 10621 | 3481104 | 17557 | 343 | 8 |
23
- | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 30630 | 11064 | 2269415 | 34147 | 2 | 24 |
24
- | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 31074 | 11064 | 3352403 | 3486 | 2 | 24 |
25
- | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 31519 | 11064 | 3352388 | 11187 | 2 | 24 |
26
-
27
- **slide\_id**: Unique identifier for the slide image, **tile\_x, tile\_y**: Coordinates of the tile within the slide at level 0, the highest pyramid level. The tiles are of size `224px X 224px`at 20x magnification (=`112um X 112um`), **level\_1 to level\_4**: Hierarchical cluster labels the tile is associated with.
28
- 2. **Visualization tool**: UMAP-based visualization of 500k curated tiles for the t1 setting (62 level_4 clusters). We provide (1) umap_metadata.csv, containing the UMAP 0 and 1 values, as well as tile metadata and (2) 500k_tiles.h5 an h5 files containing the 500k 224px X 224px tiles. Please download the 2 files for usage with out visualization tool code at [Visualization Tool Github](https://github.com/lely475/patho-ssl-data-curation/tree/main/visualization_tool)
29
 
 
 
 
30
  ## Citation
31
  Please cite our publication if you use the provided data.
32
  ```
 
1
  ---
2
  license: cc-by-4.0
 
 
3
  pretty_name: >-
4
  Revisiting Automatic Data Curation for Vision Foundation Models in Digital
5
  Pathology
 
11
 
12
  ## Data
13
  We provide the following data:
14
+ 1. **Clustering result:**
15
+ - `clustering_results/clustering_{t1,t2}.csv`: Hierarchical cluster labels for each of the 350M tiles for mode t1 and t2 as csv files. The csv contains the slide_id (WSI the tile originates from), (x,y)-coordinate of the tile at level 0 (highest pyramid level of the WSI) and cluster labels for each level (denoted as columns "level_1"..."level_4"). Then slides can be downloaded from the [TCGA](https://portal.gdc.cancer.gov/) and [GTEx](https://www.gtexportal.org/home/histologyPage) websites, for tile extraction from the WSIs we recommend [openslide](https://openslide.org/api/python/).
16
+ Structure of `clustering_results/clustering_{t1,t2}.csv`:
17
 
18
+ | slide\_id | tile\_x | tile\_y | level\_1 | level\_2 | level\_3 | level\_4 |
19
+ | ------------------------------------------------- | ------- | ------- | -------- | -------- | -------- | -------- |
20
+ | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 32406 | 10621 | 1301309 | 17404 | 2 | 24 |
21
+ | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 32850 | 10621 | 3481104 | 17557 | 343 | 8 |
22
+ | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 30630 | 11064 | 2269415 | 34147 | 2 | 24 |
23
+ | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 31074 | 11064 | 3352403 | 3486 | 2 | 24 |
24
+ | TCGA-22-1017-01Z-00-DX1.9562FE79-A261-42D3-B39... | 31519 | 11064 | 3352388 | 11187 | 2 | 24 |
 
 
 
25
 
26
+ **slide\_id**: Unique identifier for the slide image, **tile\_x, tile\_y**: Coordinates of the tile within the slide at level 0, the highest pyramid level. The tiles are of size `224px X 224px`at 20x magnification (=`112um X 112um`), **level\_1 to level\_4**: Hierarchical cluster labels the tile is associated with.
27
+ - `clustering_results/kmeans_centroids`: K-means cluster centroids at each level. This allows associating any 224x224px tiles encoded with UNI to the nearest kmeans cluster at each level.
28
+ 3. **Visualization tool**: We provide (1) UMAP coordinates for 2 million curated tiles in [metadata_N=2M.csv](https://huggingface.co/datasets/swiss-ai/patho-ssl-data-curation/blob/main/visualization_tool/metadata_N%3D2M.csv) for the t1 setting (62 level_4 clusters), and (2) tile images for a representative subset of 500k tiles, packaged in 10 tar files (`tiles_0000.tar`-`tiles_0009.tar`, each containing 224×224px pngs). (3) The file [metadata.csv](https://huggingface.co/datasets/swiss-ai/patho-ssl-data-curation/blob/main/visualization_tool/metadata.csv) contains metadata corresponding specifically to this 500k tile subset. We only provide the 500k tiles and not the full 2M due to file size limitations. Please download all files for usage with our visualization tool code at [Visualization Tool Github](https://github.com/lely475/patho-ssl-data-curation/tree/main/visualization_tool).
29
  ## Citation
30
  Please cite our publication if you use the provided data.
31
  ```