README: 25 datasets
Browse files
README.md
CHANGED
|
@@ -142,6 +142,18 @@ configs:
|
|
| 142 |
data_files:
|
| 143 |
- split: train
|
| 144 |
path: sidecar/IDRiD/*.parquet
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 145 |
- config_name: MMAC
|
| 146 |
data_files:
|
| 147 |
- split: train
|
|
@@ -292,7 +304,7 @@ configs:
|
|
| 292 |
|
| 293 |
# OphGen — curated public ophthalmic datasets
|
| 294 |
|
| 295 |
-
Curated copies of
|
| 296 |
images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
|
| 297 |
split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
|
| 298 |
|
|
@@ -324,6 +336,7 @@ In `all`, an image released by several datasets appears once per dataset (same `
|
|
| 324 |
| [GAVE2](cards/GAVE2.md) | GAVE2 | FFA, CFP | 150 | 150 | 0 | 4 |
|
| 325 |
| [GLEAM](cards/GLEAM.md) | GLEAM | IR/NIR, OCT | 1,436 | 480 | 480 | 1 |
|
| 326 |
| [IDRiD](cards/IDRiD.md) | IDRiD (segmentation and grading parts) | CFP | 382 | 42 | 107 | 2 |
|
|
|
|
| 327 |
| [MMAC](cards/MMAC.md) | MMAC 2023 (task 1) | CFP | 1,136 | 247 | 0 | 3 |
|
| 328 |
| [OCT-C8](cards/OCT-C8.md) | OCT-C8 | OCT | 6,746 | 1,051 | 1,034 | 1 |
|
| 329 |
| [OCTAVE](cards/OCTAVE.md) | OCTAVE | OCT | 2,584 | 361 | 741 | 4 |
|
|
@@ -337,7 +350,7 @@ In `all`, an image released by several datasets appears once per dataset (same `
|
|
| 337 |
| [RVO-ME](cards/RVO-ME.md) | RVO-ME | OCT | 2,057 | 229 | 580 | 3 |
|
| 338 |
| [SJTU_Peripapillary](cards/SJTU_Peripapillary.md) | SJTU peripapillary OCT | OCT | 74 | 24 | 24 | 3 |
|
| 339 |
| [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
|
| 340 |
-
| **Total** | | |
|
| 341 |
|
| 342 |
Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
|
| 343 |
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
|
|
|
|
| 142 |
data_files:
|
| 143 |
- split: train
|
| 144 |
path: sidecar/IDRiD/*.parquet
|
| 145 |
+
- config_name: MFIDDR
|
| 146 |
+
data_files:
|
| 147 |
+
- split: train
|
| 148 |
+
path: data/MFIDDR/train-*.parquet
|
| 149 |
+
- split: validation
|
| 150 |
+
path: data/MFIDDR/validation-*.parquet
|
| 151 |
+
- split: test
|
| 152 |
+
path: data/MFIDDR/test-*.parquet
|
| 153 |
+
- config_name: MFIDDR-sidecar
|
| 154 |
+
data_files:
|
| 155 |
+
- split: train
|
| 156 |
+
path: sidecar/MFIDDR/*.parquet
|
| 157 |
- config_name: MMAC
|
| 158 |
data_files:
|
| 159 |
- split: train
|
|
|
|
| 304 |
|
| 305 |
# OphGen — curated public ophthalmic datasets
|
| 306 |
|
| 307 |
+
Curated copies of 25 public ophthalmic imaging datasets, prepared for the OphGen ophthalmic generative-model project:
|
| 308 |
images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
|
| 309 |
split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
|
| 310 |
|
|
|
|
| 336 |
| [GAVE2](cards/GAVE2.md) | GAVE2 | FFA, CFP | 150 | 150 | 0 | 4 |
|
| 337 |
| [GLEAM](cards/GLEAM.md) | GLEAM | IR/NIR, OCT | 1,436 | 480 | 480 | 1 |
|
| 338 |
| [IDRiD](cards/IDRiD.md) | IDRiD (segmentation and grading parts) | CFP | 382 | 42 | 107 | 2 |
|
| 339 |
+
| [MFIDDR](cards/MFIDDR.md) | MFIDDR | CFP | 23,264 | 2,584 | 8,604 | 4 |
|
| 340 |
| [MMAC](cards/MMAC.md) | MMAC 2023 (task 1) | CFP | 1,136 | 247 | 0 | 3 |
|
| 341 |
| [OCT-C8](cards/OCT-C8.md) | OCT-C8 | OCT | 6,746 | 1,051 | 1,034 | 1 |
|
| 342 |
| [OCTAVE](cards/OCTAVE.md) | OCTAVE | OCT | 2,584 | 361 | 741 | 4 |
|
|
|
|
| 350 |
| [RVO-ME](cards/RVO-ME.md) | RVO-ME | OCT | 2,057 | 229 | 580 | 3 |
|
| 351 |
| [SJTU_Peripapillary](cards/SJTU_Peripapillary.md) | SJTU peripapillary OCT | OCT | 74 | 24 | 24 | 3 |
|
| 352 |
| [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
|
| 353 |
+
| **Total** | | | 60,232 | 8,866 | 17,784 | |
|
| 354 |
|
| 355 |
Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
|
| 356 |
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
|