README: 64 datasets
Browse files
README.md
CHANGED
|
@@ -660,6 +660,18 @@ configs:
|
|
| 660 |
data_files:
|
| 661 |
- split: train
|
| 662 |
path: sidecar/RFMiD/*.parquet
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 663 |
- config_name: ROSSA
|
| 664 |
data_files:
|
| 665 |
- split: train
|
|
@@ -728,7 +740,7 @@ configs:
|
|
| 728 |
|
| 729 |
# OphGen — curated public ophthalmic datasets
|
| 730 |
|
| 731 |
-
Curated copies of
|
| 732 |
images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
|
| 733 |
split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
|
| 734 |
|
|
@@ -806,13 +818,14 @@ In `all`, an image released by several datasets appears once per dataset (same `
|
|
| 806 |
| [REFUGE2](cards/REFUGE2.md) | REFUGE2 | CFP | 395 | 402 | 403 | 3 |
|
| 807 |
| [RETOUCH](cards/RETOUCH.md) | RETOUCH | OCT | 6,115 | 693 | 4,270 | 5 |
|
| 808 |
| [RFMiD](cards/RFMiD.md) | RFMiD 1.0 | CFP | 1,920 | 640 | 640 | 2 |
|
|
|
|
| 809 |
| [ROSSA](cards/ROSSA.md) | ROSSA | OCTA | 100 | 100 | 100 | 1 |
|
| 810 |
| [RVO-ME](cards/RVO-ME.md) | RVO-ME | OCT | 2,057 | 229 | 580 | 3 |
|
| 811 |
| [SJTU_Peripapillary](cards/SJTU_Peripapillary.md) | SJTU peripapillary OCT | OCT | 74 | 24 | 24 | 3 |
|
| 812 |
| [UMN_DME](cards/UMN_DME.md) | UMN DME | OCT | 498 | 75 | 150 | 4 |
|
| 813 |
| [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
|
| 814 |
| [UWF-Tumors](cards/UWF-Tumors.md) | UWF intraocular tumours | UWF | 1,410 | 201 | 403 | 1 |
|
| 815 |
-
| **Total** | | |
|
| 816 |
|
| 817 |
Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
|
| 818 |
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
|
|
|
|
| 660 |
data_files:
|
| 661 |
- split: train
|
| 662 |
path: sidecar/RFMiD/*.parquet
|
| 663 |
+
- config_name: RLOCT
|
| 664 |
+
data_files:
|
| 665 |
+
- split: train
|
| 666 |
+
path: data/RLOCT/train-*.parquet
|
| 667 |
+
- split: validation
|
| 668 |
+
path: data/RLOCT/validation-*.parquet
|
| 669 |
+
- split: test
|
| 670 |
+
path: data/RLOCT/test-*.parquet
|
| 671 |
+
- config_name: RLOCT-sidecar
|
| 672 |
+
data_files:
|
| 673 |
+
- split: train
|
| 674 |
+
path: sidecar/RLOCT/*.parquet
|
| 675 |
- config_name: ROSSA
|
| 676 |
data_files:
|
| 677 |
- split: train
|
|
|
|
| 740 |
|
| 741 |
# OphGen — curated public ophthalmic datasets
|
| 742 |
|
| 743 |
+
Curated copies of 64 public ophthalmic imaging datasets, prepared for the OphGen ophthalmic generative-model project:
|
| 744 |
images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
|
| 745 |
split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
|
| 746 |
|
|
|
|
| 818 |
| [REFUGE2](cards/REFUGE2.md) | REFUGE2 | CFP | 395 | 402 | 403 | 3 |
|
| 819 |
| [RETOUCH](cards/RETOUCH.md) | RETOUCH | OCT | 6,115 | 693 | 4,270 | 5 |
|
| 820 |
| [RFMiD](cards/RFMiD.md) | RFMiD 1.0 | CFP | 1,920 | 640 | 640 | 2 |
|
| 821 |
+
| [RLOCT](cards/RLOCT.md) | RLOCT (Reality Labs whole-eye OCT, anterior segment) | AS-OCT | 34,123 | 4,558 | 10,079 | 4 |
|
| 822 |
| [ROSSA](cards/ROSSA.md) | ROSSA | OCTA | 100 | 100 | 100 | 1 |
|
| 823 |
| [RVO-ME](cards/RVO-ME.md) | RVO-ME | OCT | 2,057 | 229 | 580 | 3 |
|
| 824 |
| [SJTU_Peripapillary](cards/SJTU_Peripapillary.md) | SJTU peripapillary OCT | OCT | 74 | 24 | 24 | 3 |
|
| 825 |
| [UMN_DME](cards/UMN_DME.md) | UMN DME | OCT | 498 | 75 | 150 | 4 |
|
| 826 |
| [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
|
| 827 |
| [UWF-Tumors](cards/UWF-Tumors.md) | UWF intraocular tumours | UWF | 1,410 | 201 | 403 | 1 |
|
| 828 |
+
| **Total** | | | 479,253 | 44,355 | 134,711 | |
|
| 829 |
|
| 830 |
Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
|
| 831 |
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
|