--- license: other license_name: see-original-datasets pretty_name: OphGen curated public and institutional ophthalmic datasets configs: - config_name: all default: true data_files: - split: train path: data/*/train-*.parquet - split: validation path: data/*/validation-*.parquet - split: test path: data/*/test-*.parquet - config_name: ADAM data_files: - split: train path: data/ADAM/train-*.parquet - split: validation path: data/ADAM/validation-*.parquet - split: test path: data/ADAM/test-*.parquet - config_name: ADAM-sidecar data_files: - split: train path: sidecar/ADAM/*.parquet - config_name: AGAR300 data_files: - split: train path: data/AGAR300/train-*.parquet - split: validation path: data/AGAR300/validation-*.parquet - split: test path: data/AGAR300/test-*.parquet - config_name: AIChallenger data_files: - split: train path: data/AIChallenger/train-*.parquet - split: validation path: data/AIChallenger/validation-*.parquet - split: test path: data/AIChallenger/test-*.parquet - config_name: AIChallenger-sidecar data_files: - split: train path: sidecar/AIChallenger/*.parquet - config_name: AIDK data_files: - split: train path: data/AIDK/train-*.parquet - split: validation path: data/AIDK/validation-*.parquet - split: test path: data/AIDK/test-*.parquet - config_name: AIDK-sidecar data_files: - split: train path: sidecar/AIDK/*.parquet - config_name: AMD-SD data_files: - split: train path: data/AMD-SD/train-*.parquet - split: validation path: data/AMD-SD/validation-*.parquet - split: test path: data/AMD-SD/test-*.parquet - config_name: AROI data_files: - split: train path: data/AROI/train-*.parquet - split: validation path: data/AROI/validation-*.parquet - split: test path: data/AROI/test-*.parquet - config_name: ASOCT-MSDC data_files: - split: train path: data/ASOCT-MSDC/train-*.parquet - split: validation path: data/ASOCT-MSDC/validation-*.parquet - split: test path: data/ASOCT-MSDC/test-*.parquet - config_name: ASOCT-MSDC-sidecar data_files: - split: train path: sidecar/ASOCT-MSDC/*.parquet - config_name: Birdshot-Wide data_files: - split: train path: data/Birdshot-Wide/train-*.parquet - split: validation path: data/Birdshot-Wide/validation-*.parquet - split: test path: data/Birdshot-Wide/test-*.parquet - config_name: Birdshot-Wide-sidecar data_files: - split: train path: sidecar/Birdshot-Wide/*.parquet - config_name: BUltrasound data_files: - split: train path: data/BUltrasound/train-*.parquet - split: validation path: data/BUltrasound/validation-*.parquet - split: test path: data/BUltrasound/test-*.parquet - config_name: CHAKSU data_files: - split: train path: data/CHAKSU/train-*.parquet - split: validation path: data/CHAKSU/validation-*.parquet - split: test path: data/CHAKSU/test-*.parquet - config_name: CHAKSU-sidecar data_files: - split: train path: sidecar/CHAKSU/*.parquet - config_name: chengdu_ffa data_files: - split: train path: data/chengdu_ffa/train-*.parquet - split: validation path: data/chengdu_ffa/validation-*.parquet - split: test path: data/chengdu_ffa/test-*.parquet - config_name: ChengduShiyiUBM data_files: - split: train path: data/ChengduShiyiUBM/train-*.parquet - split: validation path: data/ChengduShiyiUBM/validation-*.parquet - split: test path: data/ChengduShiyiUBM/test-*.parquet - config_name: CSDI data_files: - split: train path: data/CSDI/train-*.parquet - split: validation path: data/CSDI/validation-*.parquet - split: test path: data/CSDI/test-*.parquet - config_name: CSDI-sidecar data_files: - split: train path: sidecar/CSDI/*.parquet - config_name: CTI-FFA data_files: - split: train path: data/CTI-FFA/train-*.parquet - split: validation path: data/CTI-FFA/validation-*.parquet - split: test path: data/CTI-FFA/test-*.parquet - config_name: CTI-FFA-sidecar data_files: - split: train path: sidecar/CTI-FFA/*.parquet - config_name: CTI-ICGA data_files: - split: train path: data/CTI-ICGA/train-*.parquet - split: validation path: data/CTI-ICGA/validation-*.parquet - split: test path: data/CTI-ICGA/test-*.parquet - config_name: CTI-ICGA-sidecar data_files: - split: train path: sidecar/CTI-ICGA/*.parquet - config_name: Data_on_OCT_and_Fundus data_files: - split: train path: data/Data_on_OCT_and_Fundus/train-*.parquet - split: validation path: data/Data_on_OCT_and_Fundus/validation-*.parquet - split: test path: data/Data_on_OCT_and_Fundus/test-*.parquet - config_name: Data_on_OCT_and_Fundus-sidecar data_files: - split: train path: sidecar/Data_on_OCT_and_Fundus/*.parquet - config_name: DDR data_files: - split: train path: data/DDR/train-*.parquet - split: validation path: data/DDR/validation-*.parquet - split: test path: data/DDR/test-*.parquet - config_name: DDR-sidecar data_files: - split: train path: sidecar/DDR/*.parquet - config_name: DeepDRiD data_files: - split: train path: data/DeepDRiD/train-*.parquet - split: validation path: data/DeepDRiD/validation-*.parquet - split: test path: data/DeepDRiD/test-*.parquet - config_name: DeepDRiD-sidecar data_files: - split: train path: sidecar/DeepDRiD/*.parquet - config_name: DMR data_files: - split: train path: data/DMR/train-*.parquet - split: validation path: data/DMR/validation-*.parquet - split: test path: data/DMR/test-*.parquet - config_name: DMR-sidecar data_files: - split: train path: sidecar/DMR/*.parquet - config_name: DRIVE data_files: - split: train path: data/DRIVE/train-*.parquet - split: validation path: data/DRIVE/validation-*.parquet - split: test path: data/DRIVE/test-*.parquet - config_name: DRIVE-sidecar data_files: - split: train path: sidecar/DRIVE/*.parquet - config_name: Duke_DME data_files: - split: train path: data/Duke_DME/train-*.parquet - split: validation path: data/Duke_DME/validation-*.parquet - split: test path: data/Duke_DME/test-*.parquet - config_name: Duke_iAMD data_files: - split: train path: data/Duke_iAMD/train-*.parquet - split: validation path: data/Duke_iAMD/validation-*.parquet - split: test path: data/Duke_iAMD/test-*.parquet - config_name: Duke_iAMD-sidecar data_files: - split: train path: sidecar/Duke_iAMD/*.parquet - config_name: Duke_Srinivasan data_files: - split: train path: data/Duke_Srinivasan/train-*.parquet - split: validation path: data/Duke_Srinivasan/validation-*.parquet - split: test path: data/Duke_Srinivasan/test-*.parquet - config_name: Duke_Srinivasan-sidecar data_files: - split: train path: sidecar/Duke_Srinivasan/*.parquet - config_name: eryuanFundusT17RDCls data_files: - split: train path: data/eryuanFundusT17RDCls/train-*.parquet - split: validation path: data/eryuanFundusT17RDCls/validation-*.parquet - split: test path: data/eryuanFundusT17RDCls/test-*.parquet - config_name: eyFRDCls data_files: - split: train path: data/eyFRDCls/train-*.parquet - split: validation path: data/eyFRDCls/validation-*.parquet - split: test path: data/eyFRDCls/test-*.parquet - config_name: eyOCTset data_files: - split: train path: data/eyOCTset/train-*.parquet - split: validation path: data/eyOCTset/validation-*.parquet - split: test path: data/eyOCTset/test-*.parquet - config_name: FairSeg data_files: - split: train path: data/FairSeg/train-*.parquet - split: validation path: data/FairSeg/validation-*.parquet - split: test path: data/FairSeg/test-*.parquet - config_name: FairSeg-sidecar data_files: - split: train path: sidecar/FairSeg/*.parquet - config_name: Fang_Denoising data_files: - split: train path: data/Fang_Denoising/train-*.parquet - split: validation path: data/Fang_Denoising/validation-*.parquet - split: test path: data/Fang_Denoising/test-*.parquet - config_name: Final_eryuan_FFA data_files: - split: train path: data/Final_eryuan_FFA/train-*.parquet - split: validation path: data/Final_eryuan_FFA/validation-*.parquet - split: test path: data/Final_eryuan_FFA/test-*.parquet - config_name: FIVES data_files: - split: train path: data/FIVES/train-*.parquet - split: validation path: data/FIVES/validation-*.parquet - split: test path: data/FIVES/test-*.parquet - config_name: FIVES-sidecar data_files: - split: train path: sidecar/FIVES/*.parquet - config_name: FQ_Datasets data_files: - split: train path: data/FQ_Datasets/train-*.parquet - split: validation path: data/FQ_Datasets/validation-*.parquet - split: test path: data/FQ_Datasets/test-*.parquet - config_name: FUND data_files: - split: train path: data/FUND/train-*.parquet - split: validation path: data/FUND/validation-*.parquet - split: test path: data/FUND/test-*.parquet - config_name: FUND-sidecar data_files: - split: train path: sidecar/FUND/*.parquet - config_name: G1020 data_files: - split: train path: data/G1020/train-*.parquet - split: validation path: data/G1020/validation-*.parquet - split: test path: data/G1020/test-*.parquet - config_name: G1020-sidecar data_files: - split: train path: sidecar/G1020/*.parquet - config_name: GAVE2 data_files: - split: train path: data/GAVE2/train-*.parquet - split: validation path: data/GAVE2/validation-*.parquet - config_name: GAVE2-sidecar data_files: - split: train path: sidecar/GAVE2/*.parquet - config_name: GLEAM data_files: - split: train path: data/GLEAM/train-*.parquet - split: validation path: data/GLEAM/validation-*.parquet - split: test path: data/GLEAM/test-*.parquet - config_name: GLEAM-sidecar data_files: - split: train path: sidecar/GLEAM/*.parquet - config_name: handan_ffa data_files: - split: train path: data/handan_ffa/train-*.parquet - split: validation path: data/handan_ffa/validation-*.parquet - split: test path: data/handan_ffa/test-*.parquet - config_name: handan_ffa2 data_files: - split: train path: data/handan_ffa2/train-*.parquet - split: validation path: data/handan_ffa2/validation-*.parquet - split: test path: data/handan_ffa2/test-*.parquet - config_name: HEI-MED data_files: - split: train path: data/HEI-MED/train-*.parquet - split: validation path: data/HEI-MED/validation-*.parquet - split: test path: data/HEI-MED/test-*.parquet - config_name: HEI-MED-sidecar data_files: - split: train path: sidecar/HEI-MED/*.parquet - config_name: HYAMD data_files: - split: train path: data/HYAMD/train-*.parquet - split: validation path: data/HYAMD/validation-*.parquet - split: test path: data/HYAMD/test-*.parquet - config_name: HYAMD-sidecar data_files: - split: train path: sidecar/HYAMD/*.parquet - config_name: HYGD data_files: - split: train path: data/HYGD/train-*.parquet - split: validation path: data/HYGD/validation-*.parquet - split: test path: data/HYGD/test-*.parquet - config_name: HYGD-sidecar data_files: - split: train path: sidecar/HYGD/*.parquet - config_name: IDRiD data_files: - split: train path: data/IDRiD/train-*.parquet - split: validation path: data/IDRiD/validation-*.parquet - split: test path: data/IDRiD/test-*.parquet - config_name: IDRiD-sidecar data_files: - split: train path: sidecar/IDRiD/*.parquet - config_name: IOVS2015 data_files: - split: train path: data/IOVS2015/train-*.parquet - split: validation path: data/IOVS2015/validation-*.parquet - split: test path: data/IOVS2015/test-*.parquet - config_name: IOVS2015-sidecar data_files: - split: train path: sidecar/IOVS2015/*.parquet - config_name: JRC-FA_FAF_IR data_files: - split: train path: data/JRC-FA_FAF_IR/train-*.parquet - split: validation path: data/JRC-FA_FAF_IR/validation-*.parquet - split: test path: data/JRC-FA_FAF_IR/test-*.parquet - config_name: JSIEC data_files: - split: train path: data/JSIEC/train-*.parquet - split: validation path: data/JSIEC/validation-*.parquet - split: test path: data/JSIEC/test-*.parquet - config_name: KAP_vit data_files: - split: train path: data/KAP_vit/train-*.parquet - split: validation path: data/KAP_vit/validation-*.parquet - split: test path: data/KAP_vit/test-*.parquet - config_name: Labeled_Retinal_OCT data_files: - split: train path: data/Labeled_Retinal_OCT/train-*.parquet - split: validation path: data/Labeled_Retinal_OCT/validation-*.parquet - split: test path: data/Labeled_Retinal_OCT/test-*.parquet - config_name: Labeled_Retinal_OCT-sidecar data_files: - split: train path: sidecar/Labeled_Retinal_OCT/*.parquet - config_name: MCOA data_files: - split: train path: data/MCOA/train-*.parquet - split: validation path: data/MCOA/validation-*.parquet - split: test path: data/MCOA/test-*.parquet - config_name: MCOA-sidecar data_files: - split: train path: sidecar/MCOA/*.parquet - config_name: MFIDDR data_files: - split: train path: data/MFIDDR/train-*.parquet - split: validation path: data/MFIDDR/validation-*.parquet - split: test path: data/MFIDDR/test-*.parquet - config_name: MFIDDR-sidecar data_files: - split: train path: sidecar/MFIDDR/*.parquet - config_name: MMAC data_files: - split: train path: data/MMAC/train-*.parquet - split: validation path: data/MMAC/validation-*.parquet - config_name: MMAC-sidecar data_files: - split: train path: sidecar/MMAC/*.parquet - config_name: MMRDR data_files: - split: train path: data/MMRDR/train-*.parquet - split: validation path: data/MMRDR/validation-*.parquet - split: test path: data/MMRDR/test-*.parquet - config_name: MMRDR-sidecar data_files: - split: train path: sidecar/MMRDR/*.parquet - config_name: OCT-C8 data_files: - split: train path: data/OCT-C8/train-*.parquet - split: validation path: data/OCT-C8/validation-*.parquet - split: test path: data/OCT-C8/test-*.parquet - config_name: OCT_MS_and_HC data_files: - split: train path: data/OCT_MS_and_HC/train-*.parquet - split: validation path: data/OCT_MS_and_HC/validation-*.parquet - split: test path: data/OCT_MS_and_HC/test-*.parquet - config_name: OCT_MS_and_HC-sidecar data_files: - split: train path: sidecar/OCT_MS_and_HC/*.parquet - config_name: OCTA-25K data_files: - split: train path: data/OCTA-25K/train-*.parquet - split: validation path: data/OCTA-25K/validation-*.parquet - split: test path: data/OCTA-25K/test-*.parquet - config_name: OCTA500 data_files: - split: train path: data/OCTA500/train-*.parquet - split: validation path: data/OCTA500/validation-*.parquet - split: test path: data/OCTA500/test-*.parquet - config_name: OCTA500-sidecar data_files: - split: train path: sidecar/OCTA500/*.parquet - config_name: OCTAVE data_files: - split: train path: data/OCTAVE/train-*.parquet - split: validation path: data/OCTAVE/validation-*.parquet - split: test path: data/OCTAVE/test-*.parquet - config_name: OCTAVE-sidecar data_files: - split: train path: sidecar/OCTAVE/*.parquet - config_name: OCTDL data_files: - split: train path: data/OCTDL/train-*.parquet - split: validation path: data/OCTDL/validation-*.parquet - split: test path: data/OCTDL/test-*.parquet - config_name: OCTDL-sidecar data_files: - split: train path: sidecar/OCTDL/*.parquet - config_name: OCTID data_files: - split: train path: data/OCTID/train-*.parquet - split: validation path: data/OCTID/validation-*.parquet - split: test path: data/OCTID/test-*.parquet - config_name: OCTID-sidecar data_files: - split: train path: sidecar/OCTID/*.parquet - config_name: Oculo data_files: - split: train path: data/Oculo/train-*.parquet - split: validation path: data/Oculo/validation-*.parquet - split: test path: data/Oculo/test-*.parquet - config_name: Oculo-sidecar data_files: - split: train path: sidecar/Oculo/*.parquet - config_name: ODIR data_files: - split: train path: data/ODIR/train-*.parquet - split: validation path: data/ODIR/validation-*.parquet - split: test path: data/ODIR/test-*.parquet - config_name: ODIR-sidecar data_files: - split: train path: sidecar/ODIR/*.parquet - config_name: OIMHS data_files: - split: train path: data/OIMHS/train-*.parquet - split: validation path: data/OIMHS/validation-*.parquet - split: test path: data/OIMHS/test-*.parquet - config_name: OIMHS-sidecar data_files: - split: train path: sidecar/OIMHS/*.parquet - config_name: OLIVES data_files: - split: train path: data/OLIVES/train-*.parquet - split: validation path: data/OLIVES/validation-*.parquet - split: test path: data/OLIVES/test-*.parquet - config_name: OLIVES-sidecar data_files: - split: train path: sidecar/OLIVES/*.parquet - config_name: ORIGA data_files: - split: train path: data/ORIGA/train-*.parquet - split: validation path: data/ORIGA/validation-*.parquet - split: test path: data/ORIGA/test-*.parquet - config_name: ORIGA-sidecar data_files: - split: train path: sidecar/ORIGA/*.parquet - config_name: OUWFD data_files: - split: train path: data/OUWFD/train-*.parquet - split: validation path: data/OUWFD/validation-*.parquet - split: test path: data/OUWFD/test-*.parquet - config_name: OUWFD-sidecar data_files: - split: train path: sidecar/OUWFD/*.parquet - config_name: PALM data_files: - split: train path: data/PALM/train-*.parquet - split: validation path: data/PALM/validation-*.parquet - split: test path: data/PALM/test-*.parquet - config_name: PALM-sidecar data_files: - split: train path: sidecar/PALM/*.parquet - config_name: PAPILA data_files: - split: train path: data/PAPILA/train-*.parquet - split: validation path: data/PAPILA/validation-*.parquet - split: test path: data/PAPILA/test-*.parquet - config_name: PAPILA-sidecar data_files: - split: train path: sidecar/PAPILA/*.parquet - config_name: PKU37 data_files: - split: train path: data/PKU37/train-*.parquet - split: validation path: data/PKU37/validation-*.parquet - split: test path: data/PKU37/test-*.parquet - config_name: PRIME-FP20 data_files: - split: train path: data/PRIME-FP20/train-*.parquet - split: validation path: data/PRIME-FP20/validation-*.parquet - split: test path: data/PRIME-FP20/test-*.parquet - config_name: PRIME-FP20-sidecar data_files: - split: train path: sidecar/PRIME-FP20/*.parquet - config_name: pureCFP55w data_files: - split: train path: data/pureCFP55w/train-*.parquet - split: validation path: data/pureCFP55w/validation-*.parquet - split: test path: data/pureCFP55w/test-*.parquet - config_name: Rare_Retinal_OCT data_files: - split: train path: data/Rare_Retinal_OCT/train-*.parquet - split: validation path: data/Rare_Retinal_OCT/validation-*.parquet - split: test path: data/Rare_Retinal_OCT/test-*.parquet - config_name: Rare_Retinal_OCT-sidecar data_files: - split: train path: sidecar/Rare_Retinal_OCT/*.parquet - config_name: RAVIR data_files: - split: train path: data/RAVIR/train-*.parquet - split: validation path: data/RAVIR/validation-*.parquet - split: test path: data/RAVIR/test-*.parquet - config_name: RAVIR-sidecar data_files: - split: train path: sidecar/RAVIR/*.parquet - config_name: REFUGE2 data_files: - split: train path: data/REFUGE2/train-*.parquet - split: validation path: data/REFUGE2/validation-*.parquet - split: test path: data/REFUGE2/test-*.parquet - config_name: REFUGE2-sidecar data_files: - split: train path: sidecar/REFUGE2/*.parquet - config_name: RETOUCH data_files: - split: train path: data/RETOUCH/train-*.parquet - split: validation path: data/RETOUCH/validation-*.parquet - split: test path: data/RETOUCH/test-*.parquet - config_name: RETOUCH-sidecar data_files: - split: train path: sidecar/RETOUCH/*.parquet - config_name: RFMiD data_files: - split: train path: data/RFMiD/train-*.parquet - split: validation path: data/RFMiD/validation-*.parquet - split: test path: data/RFMiD/test-*.parquet - config_name: RFMiD-sidecar data_files: - split: train path: sidecar/RFMiD/*.parquet - config_name: RLOCT data_files: - split: train path: data/RLOCT/train-*.parquet - split: validation path: data/RLOCT/validation-*.parquet - split: test path: data/RLOCT/test-*.parquet - config_name: RLOCT-sidecar data_files: - split: train path: sidecar/RLOCT/*.parquet - config_name: ROSSA data_files: - split: train path: data/ROSSA/train-*.parquet - split: validation path: data/ROSSA/validation-*.parquet - split: test path: data/ROSSA/test-*.parquet - config_name: ROSSA-sidecar data_files: - split: train path: sidecar/ROSSA/*.parquet - config_name: RVO-ME data_files: - split: train path: data/RVO-ME/train-*.parquet - split: validation path: data/RVO-ME/validation-*.parquet - split: test path: data/RVO-ME/test-*.parquet - config_name: RVO-ME-sidecar data_files: - split: train path: sidecar/RVO-ME/*.parquet - config_name: SJTU_Peripapillary data_files: - split: train path: data/SJTU_Peripapillary/train-*.parquet - split: validation path: data/SJTU_Peripapillary/validation-*.parquet - split: test path: data/SJTU_Peripapillary/test-*.parquet - config_name: SLID data_files: - split: train path: data/SLID/train-*.parquet - split: validation path: data/SLID/validation-*.parquet - split: test path: data/SLID/test-*.parquet - config_name: tongren16G95DFundus data_files: - split: train path: data/tongren16G95DFundus/train-*.parquet - split: validation path: data/tongren16G95DFundus/validation-*.parquet - split: test path: data/tongren16G95DFundus/test-*.parquet - config_name: tongren16G95DFundus-sidecar data_files: - split: train path: sidecar/tongren16G95DFundus/*.parquet - config_name: tongren_archives data_files: - split: train path: data/tongren_archives/train-*.parquet - split: validation path: data/tongren_archives/validation-*.parquet - split: test path: data/tongren_archives/test-*.parquet - config_name: tongrenHandan data_files: - split: train path: data/tongrenHandan/train-*.parquet - split: validation path: data/tongrenHandan/validation-*.parquet - split: test path: data/tongrenHandan/test-*.parquet - config_name: trhd3_FRD_Cls66_Train data_files: - split: train path: data/trhd3_FRD_Cls66_Train/train-*.parquet - split: validation path: data/trhd3_FRD_Cls66_Train/validation-*.parquet - split: test path: data/trhd3_FRD_Cls66_Train/test-*.parquet - config_name: UESTC_OCT data_files: - split: train path: data/UESTC_OCT/train-*.parquet - split: validation path: data/UESTC_OCT/validation-*.parquet - split: test path: data/UESTC_OCT/test-*.parquet - config_name: UMN_DME data_files: - split: train path: data/UMN_DME/train-*.parquet - split: validation path: data/UMN_DME/validation-*.parquet - split: test path: data/UMN_DME/test-*.parquet - config_name: UWF-DR_2026 data_files: - split: train path: data/UWF-DR_2026/train-*.parquet - split: validation path: data/UWF-DR_2026/validation-*.parquet - split: test path: data/UWF-DR_2026/test-*.parquet - config_name: UWF-DR_2026-sidecar data_files: - split: train path: sidecar/UWF-DR_2026/*.parquet - config_name: UWF-Tumors data_files: - split: train path: data/UWF-Tumors/train-*.parquet - split: validation path: data/UWF-Tumors/validation-*.parquet - split: test path: data/UWF-Tumors/test-*.parquet - config_name: UWF-Tumors-sidecar data_files: - split: train path: sidecar/UWF-Tumors/*.parquet - config_name: yinhai_ffa data_files: - split: train path: data/yinhai_ffa/train-*.parquet - split: validation path: data/yinhai_ffa/validation-*.parquet - split: test path: data/yinhai_ffa/test-*.parquet - config_name: yinhai_uwf data_files: - split: train path: data/yinhai_uwf/train-*.parquet - split: validation path: data/yinhai_uwf/validation-*.parquet - split: test path: data/yinhai_uwf/test-*.parquet - config_name: YinhaiUBM data_files: - split: train path: data/YinhaiUBM/train-*.parquet - split: validation path: data/YinhaiUBM/validation-*.parquet - split: test path: data/YinhaiUBM/test-*.parquet --- # OphGen — curated public and institutional ophthalmic datasets Curated copies of 89 public and institutional ophthalmic imaging datasets, prepared for the OphGen ophthalmic generative-model project: images as stored in the OphGen image store (de-identified, de-duplicated, native resolution), the train / validation / test split used in the project, six captions per image (short / medium / long × English / Chinese) and all original task annotations. **Public datasets: use and licence follow each original dataset** (see the dataset cards). Please cite the original publications and respect their terms. **Institutional datasets (private hospital data): access is restricted to the OphGen project; do not redistribute or re-upload.** Images were de-identified (burned-in personal text removed); their `source_paths` are paths in the hospital archives. ## Loading ```python from datasets import load_dataset ds = load_dataset("Kaphathy/OphGen", "REFUGE2") # one dataset, its own splits and labels side = load_dataset("Kaphathy/OphGen", "REFUGE2-sidecar") # its other annotation files (path, size, bytes), if any pre = load_dataset("Kaphathy/OphGen", "all") # every dataset (default config), e.g. for pretraining ``` In `all`, an image released by several datasets appears once per dataset (same `image_id`, same split); deduplicate on `image_id` if needed. ## Datasets | Config | Dataset | Type | Modalities | Train | Validation | Test | Tasks | | :--- | :--- | :--- | :--- | ---: | ---: | ---: | ---: | | [ADAM](cards/ADAM.md) | ADAM (iChallenge-AMD official training set) | public | CFP | 286 | 66 | 45 | 1 | | [AGAR300](cards/AGAR300.md) | AGAR300 | public | CFP | 25 | 2 | 1 | 0 | | [AIChallenger](cards/AIChallenger.md) | AI Challenger 2018 retinal edema | public | OCT | 8,960 | 1,920 | 1,920 | 1 | | [AIDK](cards/AIDK.md) | AIDK | public | AS-OCT | 818 | 116 | 234 | 1 | | [AMD-SD](cards/AMD-SD.md) | AMD-SD | public | OCT | 2,121 | 296 | 613 | 5 | | [AROI](cards/AROI.md) | AROI | public | OCT | 768 | 137 | 231 | 5 | | [ASOCT-MSDC](cards/ASOCT-MSDC.md) | ASOCT-MSDC | public | AS-OCT | 773 | 110 | 221 | 5 | | [Birdshot-Wide](cards/Birdshot-Wide.md) | Birdshot-Wide | public | UWF | 3,865 | 553 | 1,104 | 2 | | [BUltrasound](cards/BUltrasound.md) | BUltrasound | institutional | Conventional ocular B-scan ultrasonography | 101,473 | 5,689 | 5,574 | 0 | | [CHAKSU](cards/CHAKSU.md) | Chákṣu | public | CFP | 903 | 100 | 335 | 3 | | [chengdu_ffa](cards/chengdu_ffa.md) | chengdu_ffa | institutional | FFA | 3,519 | 473 | 934 | 1 | | [ChengduShiyiUBM](cards/ChengduShiyiUBM.md) | ChengduShiyiUBM | institutional | UBM | 413,564 | 22,972 | 22,970 | 0 | | [CSDI](cards/CSDI.md) | CSDI | public | CFP | 131 | 19 | 37 | 4 | | [CTI-FFA](cards/CTI-FFA.md) | CTI (FFA part) | public | FFA | 1,244 | 138 | 174 | 3 | | [CTI-ICGA](cards/CTI-ICGA.md) | CTI (ICGA part) | public | ICGA | 1,154 | 129 | 174 | 3 | | [Data_on_OCT_and_Fundus](cards/Data_on_OCT_and_Fundus.md) | AFIO | public | OCT, CFP | 70 | 12 | 16 | 4 | | [DDR](cards/DDR.md) | DDR | public | CFP | 6,552 | 2,688 | 4,232 | 3 | | [DeepDRiD](cards/DeepDRiD.md) | DeepDRiD | public | CFP, UWF | 1,329 | 446 | 450 | 3 | | [DMR](cards/DMR.md) | DMR (Davis grading) | public | CFP | 6,960 | 992 | 1,987 | 3 | | [DRIVE](cards/DRIVE.md) | DRIVE | public | CFP | 18 | 2 | 20 | 1 | | [Duke_DME](cards/Duke_DME.md) | Duke DME (Chiu) | public | OCT | 77 | 11 | 22 | 3 | | [Duke_iAMD](cards/Duke_iAMD.md) | Duke iAMD | public | OCT | 26,840 | 3,799 | 7,662 | 3 | | [Duke_Srinivasan](cards/Duke_Srinivasan.md) | Duke OCT (Srinivasan) | public | OCT | 2,263 | 299 | 669 | 2 | | [eryuanFundusT17RDCls](cards/eryuanFundusT17RDCls.md) | eryuanFundusT17RDCls | institutional | CFP | 13,115 | 1,906 | 3,719 | 1 | | [eyFRDCls](cards/eyFRDCls.md) | eyFRDCls | institutional | CFP | 418 | 53 | 139 | 1 | | [eyOCTset](cards/eyOCTset.md) | eyOCTset | institutional | OCT | 160,167 | 8,897 | 8,899 | 0 | | [FairSeg](cards/FairSeg.md) | Harvard-FairSeg | public | IR/NIR | 7,200 | 800 | 2,000 | 2 | | [Fang_Denoising](cards/Fang_Denoising.md) | Duke OCT denoising (Fang) | public | OCT | 24 | 4 | 6 | 2 | | [Final_eryuan_FFA](cards/Final_eryuan_FFA.md) | Final_eryuan_FFA | institutional | FFA | 150,098 | 8,374 | 8,562 | 0 | | [FIVES](cards/FIVES.md) | FIVES | public | CFP | 536 | 59 | 199 | 2 | | [FQ_Datasets](cards/FQ_Datasets.md) | FQ_Datasets | institutional | CFP | 2,681 | 402 | 802 | 1 | | [FUND](cards/FUND.md) | FUND | public | OCT, CFP | 1,173 | 277 | 293 | 3 | | [G1020](cards/G1020.md) | G1020 | public | CFP | 713 | 102 | 204 | 2 | | [GAVE2](cards/GAVE2.md) | GAVE2 | public | FFA, CFP | 150 | 150 | 0 | 4 | | [GLEAM](cards/GLEAM.md) | GLEAM | public | IR/NIR, OCT | 1,436 | 480 | 480 | 1 | | [handan_ffa](cards/handan_ffa.md) | handan_ffa | institutional | FFA, CFP | 26,560 | 1,478 | 1,471 | 0 | | [handan_ffa2](cards/handan_ffa2.md) | handan_ffa2 | institutional | FFA | 17,465 | 980 | 991 | 0 | | [HEI-MED](cards/HEI-MED.md) | HEI-MED | public | CFP | 118 | 17 | 34 | 6 | | [HYAMD](cards/HYAMD.md) | HYAMD | public | CFP | 1,084 | 164 | 312 | 3 | | [HYGD](cards/HYGD.md) | HYGD | public | CFP | 515 | 75 | 147 | 2 | | [IDRiD](cards/IDRiD.md) | IDRiD (segmentation and grading parts) | public | CFP | 382 | 42 | 107 | 2 | | [IOVS2015](cards/IOVS2015.md) | IOVS2015 (Duke DME leakage) | public | FFA | 401 | 57 | 115 | 3 | | [JRC-FA_FAF_IR](cards/JRC-FA_FAF_IR.md) | JRC | public | FFA, FAF, IR/NIR | 84 | 12 | 24 | 1 | | [JSIEC](cards/JSIEC.md) | JSIEC | public | CFP | 697 | 100 | 199 | 1 | | [KAP_vit](cards/KAP_vit.md) | KAP_vit | institutional | CFP | 234,828 | 23,034 | 39,425 | 1 | | [Labeled_Retinal_OCT](cards/Labeled_Retinal_OCT.md) | NEH OCT | public | OCT | 11,686 | 1,665 | 3,334 | 4 | | [MCOA](cards/MCOA.md) | MCOA | public | AS-OCT, Slit-lamp anterior-segment photography | 4,627 | 646 | 1,326 | 6 | | [MFIDDR](cards/MFIDDR.md) | MFIDDR | public | CFP | 23,264 | 2,584 | 8,604 | 4 | | [MMAC](cards/MMAC.md) | MMAC 2023 (task 1) | public | CFP | 1,136 | 247 | 0 | 3 | | [MMRDR](cards/MMRDR.md) | MMRDR | public | CFP, UWF, OCT | 17,075 | 1,897 | 5,380 | 4 | | [OCT-C8](cards/OCT-C8.md) | OCT-C8 | public | OCT | 6,746 | 1,051 | 1,034 | 1 | | [OCT_MS_and_HC](cards/OCT_MS_and_HC.md) | JHU OCT | public | OCT | 1,176 | 196 | 343 | 5 | | [OCTA-25K](cards/OCTA-25K.md) | OCTA-25K | public | OCTA | 12,505 | 695 | 695 | 0 | | [OCTA500](cards/OCTA500.md) | OCTA-500 | public | OCTA, OCT | 277,645 | 14,140 | 70,700 | 4 | | [OCTAVE](cards/OCTAVE.md) | OCTAVE | public | OCT | 2,584 | 361 | 741 | 4 | | [OCTDL](cards/OCTDL.md) | OCTDL | public | OCT | 1,406 | 201 | 402 | 4 | | [OCTID](cards/OCTID.md) | OCTID | public | OCT | 399 | 60 | 113 | 2 | | [Oculo](cards/Oculo.md) | Oculo | public | Conventional ocular B-scan ultrasonography | 1,189 | 162 | 245 | 2 | | [ODIR](cards/ODIR.md) | ODIR-5K | public | CFP | 6,215 | 701 | 1,001 | 4 | | [OIMHS](cards/OIMHS.md) | OIMHS | public | OCT | 2,697 | 389 | 773 | 5 | | [OLIVES](cards/OLIVES.md) | OLIVES | public | OCT, IR/NIR | 104,590 | 14,305 | 29,361 | 8 | | [ORIGA](cards/ORIGA.md) | ORIGA | public | CFP | 293 | 32 | 325 | 4 | | [OUWFD](cards/OUWFD.md) | OUWFD | public | UWF | 490 | 70 | 140 | 5 | | [PALM](cards/PALM.md) | PALM | public | CFP | 386 | 401 | 397 | 3 | | [PAPILA](cards/PAPILA.md) | PAPILA | public | CFP | 342 | 48 | 98 | 6 | | [PKU37](cards/PKU37.md) | PKU37 | public | OCT | 1,211 | 147 | 370 | 2 | | [PRIME-FP20](cards/PRIME-FP20.md) | PRIME-FP20 | public | FFA, UWF | 20 | 4 | 6 | 3 | | [pureCFP55w](cards/pureCFP55w.md) | pureCFP55w | institutional | CFP | 485,999 | 27,000 | 27,000 | 0 | | [Rare_Retinal_OCT](cards/Rare_Retinal_OCT.md) | Rare retinal disease OCT images | public | OCT | 134 | 19 | 38 | 2 | | [RAVIR](cards/RAVIR.md) | RAVIR | public | IR/NIR | 20 | 2 | 19 | 2 | | [REFUGE2](cards/REFUGE2.md) | REFUGE2 | public | CFP | 395 | 402 | 403 | 3 | | [RETOUCH](cards/RETOUCH.md) | RETOUCH | public | OCT | 6,115 | 693 | 4,270 | 5 | | [RFMiD](cards/RFMiD.md) | RFMiD 1.0 | public | CFP | 1,920 | 640 | 640 | 2 | | [RLOCT](cards/RLOCT.md) | RLOCT (Reality Labs whole-eye OCT, anterior segment) | public | AS-OCT | 34,123 | 4,558 | 10,079 | 4 | | [ROSSA](cards/ROSSA.md) | ROSSA | public | OCTA | 100 | 100 | 100 | 1 | | [RVO-ME](cards/RVO-ME.md) | RVO-ME | public | OCT | 2,057 | 229 | 580 | 3 | | [SJTU_Peripapillary](cards/SJTU_Peripapillary.md) | SJTU peripapillary OCT | public | OCT | 74 | 24 | 24 | 3 | | [SLID](cards/SLID.md) | SLID | public | Slit-lamp anterior-segment photography | 2,355 | 131 | 131 | 0 | | [tongren16G95DFundus](cards/tongren16G95DFundus.md) | tongren16G95DFundus | institutional | CFP | 172,345 | 24,419 | 49,171 | 1 | | [tongren_archives](cards/tongren_archives.md) | tongren_archives | institutional | CFP, External ocular photography | 674,175 | 58,101 | 91,359 | 0 | | [tongrenHandan](cards/tongrenHandan.md) | tongrenHandan | institutional | CFP | 7,012 | 987 | 1,974 | 1 | | [trhd3_FRD_Cls66_Train](cards/trhd3_FRD_Cls66_Train.md) | trhd3_FRD_Cls66_Train | institutional | CFP | 7,780 | 1,106 | 2,176 | 1 | | [UESTC_OCT](cards/UESTC_OCT.md) | UESTC OCT-R1/R2 | public | OCT | 29,184 | 3,072 | 2,000 | 1 | | [UMN_DME](cards/UMN_DME.md) | UMN DME | public | OCT | 498 | 75 | 150 | 4 | | [UWF-DR_2026](cards/UWF-DR_2026.md) | UWF-DR | public | UWF | 1,135 | 158 | 327 | 2 | | [UWF-Tumors](cards/UWF-Tumors.md) | UWF intraocular tumours | public | UWF | 1,410 | 201 | 403 | 1 | | [yinhai_ffa](cards/yinhai_ffa.md) | yinhai_ffa | institutional | FFA | 812 | 34 | 52 | 0 | | [yinhai_uwf](cards/yinhai_uwf.md) | yinhai_uwf | institutional | UWF | 175,209 | 9,684 | 9,790 | 0 | | [YinhaiUBM](cards/YinhaiUBM.md) | YinhaiUBM | institutional | UBM | 75,720 | 4,012 | 3,857 | 0 | | **Total** | | | | 3,359,812 | 264,081 | 447,684 | | Split rules: `official` = official split; `official_val_from_train` = 10% of the official training set (by patient) used as 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 for images without task labels. An image with different splits across tasks takes the strictest (test > validation > train). ## Columns (dataset configs and `all`) | Column | Description | | :--- | :--- | | `image_id` | SHA-256 of the original file (stable identifier) | | `image` | image bytes in the stored format; native resolution (no resizing) | | `split`, `split_rule` | partition used in the project and how it was assigned | | `modality`, `modality_name` | one of 13 standard ophthalmic modalities | | `width`, `height`, `format`, `field_of_view`, `device`, `eye`, `sex`, `age` | metadata (empty when not documented) | | `representation`, `octa_layer`, `scan_area_mm` | OCTA-specific fields | | `short_en` … `long_zh` | captions built only from recorded facts; images in official test sets carry no task labels in captions (`benchmark_holdout`) | | `source_paths` | original path(s) of the image in the official release (public) or hospital archive (institutional) | | `labels` | task labels of this dataset: `task`, `name` (label name, value encoding included), `value`, `num`, `official_split`, `annotation` / `label_file` (path of the annotation, embedded in `masks` or stored in the `-sidecar` config) | | `masks` | raster annotation masks for this image (`name` matches `labels.name`, `path` matches `labels.annotation`) | `-sidecar` configs: `path`, `size`, `bytes` of every other annotation file of the release (tables, JSON, NPZ, MAT, XML …); write `bytes` to `path` to restore the original layout. ## Processing Exact and pixel-identical duplicates were merged (labels kept); images with burned-in personal text or non-removable overlays, undecodable or blank files were removed; images were not resized. Where the release did not document a mask value mapping, the mapping was inferred and is stated in the label name. Full methods: OphGen supplementary methods.