README: 68 datasets
Browse files
README.md
CHANGED
|
@@ -456,6 +456,14 @@ configs:
|
|
| 456 |
data_files:
|
| 457 |
- split: train
|
| 458 |
path: sidecar/OCT_MS_and_HC/*.parquet
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 459 |
- config_name: OCTA500
|
| 460 |
data_files:
|
| 461 |
- split: train
|
|
@@ -768,7 +776,7 @@ configs:
|
|
| 768 |
|
| 769 |
# OphGen — curated public ophthalmic datasets
|
| 770 |
|
| 771 |
-
Curated copies of
|
| 772 |
images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
|
| 773 |
split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
|
| 774 |
|
|
@@ -829,6 +837,7 @@ In `all`, an image released by several datasets appears once per dataset (same `
|
|
| 829 |
| [MMRDR](cards/MMRDR.md) | MMRDR | CFP, UWF, OCT | 17,075 | 1,897 | 5,380 | 4 |
|
| 830 |
| [OCT-C8](cards/OCT-C8.md) | OCT-C8 | OCT | 6,746 | 1,051 | 1,034 | 1 |
|
| 831 |
| [OCT_MS_and_HC](cards/OCT_MS_and_HC.md) | JHU OCT | OCT | 1,176 | 196 | 343 | 5 |
|
|
|
|
| 832 |
| [OCTA500](cards/OCTA500.md) | OCTA-500 | OCTA, OCT | 277,645 | 14,140 | 70,700 | 4 |
|
| 833 |
| [OCTAVE](cards/OCTAVE.md) | OCTAVE | OCT | 2,584 | 361 | 741 | 4 |
|
| 834 |
| [OCTDL](cards/OCTDL.md) | OCTDL | OCT | 1,406 | 201 | 402 | 4 |
|
|
@@ -856,7 +865,7 @@ In `all`, an image released by several datasets appears once per dataset (same `
|
|
| 856 |
| [UMN_DME](cards/UMN_DME.md) | UMN DME | OCT | 498 | 75 | 150 | 4 |
|
| 857 |
| [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
|
| 858 |
| [UWF-Tumors](cards/UWF-Tumors.md) | UWF intraocular tumours | UWF | 1,410 | 201 | 403 | 1 |
|
| 859 |
-
| **Total** | | |
|
| 860 |
|
| 861 |
Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
|
| 862 |
validation; `labeled_7_1_2` = patient-level 7:1:2 split with a fixed seed (no official split); `pretrain_90_5_5` = 90:5:5 split
|
|
|
|
| 456 |
data_files:
|
| 457 |
- split: train
|
| 458 |
path: sidecar/OCT_MS_and_HC/*.parquet
|
| 459 |
+
- config_name: OCTA-25K
|
| 460 |
+
data_files:
|
| 461 |
+
- split: train
|
| 462 |
+
path: data/OCTA-25K/train-*.parquet
|
| 463 |
+
- split: validation
|
| 464 |
+
path: data/OCTA-25K/validation-*.parquet
|
| 465 |
+
- split: test
|
| 466 |
+
path: data/OCTA-25K/test-*.parquet
|
| 467 |
- config_name: OCTA500
|
| 468 |
data_files:
|
| 469 |
- split: train
|
|
|
|
| 776 |
|
| 777 |
# OphGen — curated public ophthalmic datasets
|
| 778 |
|
| 779 |
+
Curated copies of 68 public ophthalmic imaging datasets, prepared for the OphGen ophthalmic generative-model project:
|
| 780 |
images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
|
| 781 |
split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
|
| 782 |
|
|
|
|
| 837 |
| [MMRDR](cards/MMRDR.md) | MMRDR | CFP, UWF, OCT | 17,075 | 1,897 | 5,380 | 4 |
|
| 838 |
| [OCT-C8](cards/OCT-C8.md) | OCT-C8 | OCT | 6,746 | 1,051 | 1,034 | 1 |
|
| 839 |
| [OCT_MS_and_HC](cards/OCT_MS_and_HC.md) | JHU OCT | OCT | 1,176 | 196 | 343 | 5 |
|
| 840 |
+
| [OCTA-25K](cards/OCTA-25K.md) | OCTA-25K | OCTA | 12,505 | 695 | 695 | 0 |
|
| 841 |
| [OCTA500](cards/OCTA500.md) | OCTA-500 | OCTA, OCT | 277,645 | 14,140 | 70,700 | 4 |
|
| 842 |
| [OCTAVE](cards/OCTAVE.md) | OCTAVE | OCT | 2,584 | 361 | 741 | 4 |
|
| 843 |
| [OCTDL](cards/OCTDL.md) | OCTDL | OCT | 1,406 | 201 | 402 | 4 |
|
|
|
|
| 865 |
| [UMN_DME](cards/UMN_DME.md) | UMN DME | OCT | 498 | 75 | 150 | 4 |
|
| 866 |
| [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
|
| 867 |
| [UWF-Tumors](cards/UWF-Tumors.md) | UWF intraocular tumours | UWF | 1,410 | 201 | 403 | 1 |
|
| 868 |
+
| **Total** | | | 598,728 | 59,488 | 164,899 | |
|
| 869 |
|
| 870 |
Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
|
| 871 |
validation; `labeled_7_1_2` = patient-level 7:1:2 split with a fixed seed (no official split); `pretrain_90_5_5` = 90:5:5 split
|