Kaphathy commited on
Commit
6714a1d
·
verified ·
1 Parent(s): e5c9893

README: 52 datasets

Browse files
Files changed (1) hide show
  1. README.md +15 -2
README.md CHANGED
@@ -168,6 +168,18 @@ configs:
168
  path: data/Duke_DME/validation-*.parquet
169
  - split: test
170
  path: data/Duke_DME/test-*.parquet
 
 
 
 
 
 
 
 
 
 
 
 
171
  - config_name: FairSeg
172
  data_files:
173
  - split: train
@@ -584,7 +596,7 @@ configs:
584
 
585
  # OphGen — curated public ophthalmic datasets
586
 
587
- Curated copies of 51 public ophthalmic imaging datasets, prepared for the OphGen ophthalmic generative-model project:
588
  images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
589
  split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
590
 
@@ -619,6 +631,7 @@ In `all`, an image released by several datasets appears once per dataset (same `
619
  | [DMR](cards/DMR.md) | DMR (Davis grading) | CFP | 6,960 | 992 | 1,987 | 3 |
620
  | [DRIVE](cards/DRIVE.md) | DRIVE | CFP | 18 | 2 | 20 | 1 |
621
  | [Duke_DME](cards/Duke_DME.md) | Duke DME (Chiu) | OCT | 77 | 11 | 22 | 3 |
 
622
  | [FairSeg](cards/FairSeg.md) | Harvard-FairSeg | IR/NIR | 7,200 | 800 | 2,000 | 2 |
623
  | [Fang_Denoising](cards/Fang_Denoising.md) | Duke OCT denoising (Fang) | OCT | 24 | 4 | 6 | 2 |
624
  | [FIVES](cards/FIVES.md) | FIVES | CFP | 536 | 59 | 199 | 2 |
@@ -656,7 +669,7 @@ In `all`, an image released by several datasets appears once per dataset (same `
656
  | [SJTU_Peripapillary](cards/SJTU_Peripapillary.md) | SJTU peripapillary OCT | OCT | 74 | 24 | 24 | 3 |
657
  | [UMN_DME](cards/UMN_DME.md) | UMN DME | OCT | 498 | 75 | 150 | 4 |
658
  | [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
659
- | **Total** | | | 382,522 | 31,100 | 104,201 | |
660
 
661
  Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
662
  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
 
168
  path: data/Duke_DME/validation-*.parquet
169
  - split: test
170
  path: data/Duke_DME/test-*.parquet
171
+ - config_name: Duke_Srinivasan
172
+ data_files:
173
+ - split: train
174
+ path: data/Duke_Srinivasan/train-*.parquet
175
+ - split: validation
176
+ path: data/Duke_Srinivasan/validation-*.parquet
177
+ - split: test
178
+ path: data/Duke_Srinivasan/test-*.parquet
179
+ - config_name: Duke_Srinivasan-sidecar
180
+ data_files:
181
+ - split: train
182
+ path: sidecar/Duke_Srinivasan/*.parquet
183
  - config_name: FairSeg
184
  data_files:
185
  - split: train
 
596
 
597
  # OphGen — curated public ophthalmic datasets
598
 
599
+ Curated copies of 52 public ophthalmic imaging datasets, prepared for the OphGen ophthalmic generative-model project:
600
  images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test
601
  split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations.
602
 
 
631
  | [DMR](cards/DMR.md) | DMR (Davis grading) | CFP | 6,960 | 992 | 1,987 | 3 |
632
  | [DRIVE](cards/DRIVE.md) | DRIVE | CFP | 18 | 2 | 20 | 1 |
633
  | [Duke_DME](cards/Duke_DME.md) | Duke DME (Chiu) | OCT | 77 | 11 | 22 | 3 |
634
+ | [Duke_Srinivasan](cards/Duke_Srinivasan.md) | Duke OCT (Srinivasan) | OCT | 2,263 | 299 | 669 | 2 |
635
  | [FairSeg](cards/FairSeg.md) | Harvard-FairSeg | IR/NIR | 7,200 | 800 | 2,000 | 2 |
636
  | [Fang_Denoising](cards/Fang_Denoising.md) | Duke OCT denoising (Fang) | OCT | 24 | 4 | 6 | 2 |
637
  | [FIVES](cards/FIVES.md) | FIVES | CFP | 536 | 59 | 199 | 2 |
 
669
  | [SJTU_Peripapillary](cards/SJTU_Peripapillary.md) | SJTU peripapillary OCT | OCT | 74 | 24 | 24 | 3 |
670
  | [UMN_DME](cards/UMN_DME.md) | UMN DME | OCT | 498 | 75 | 150 | 4 |
671
  | [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | UWF | 1,135 | 158 | 327 | 2 |
672
+ | **Total** | | | 384,785 | 31,399 | 104,870 | |
673
 
674
  Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as
675
  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