{"success":true,"data":{"author":"US FDA Division of Imaging, Diagnostics, and Software Reliability (Kim, Saharkhiz, Sizikova, Lago, Sahiner, Delfino, Badano)","bucketAccessType":"public","bucketName":"didsr/ssynth_data","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-14T14:15:11.926Z","description":"S-SYNTH is a knowledge-based, open-source skin simulation framework and pre-generated dataset from the US Food and Drug Administration's Division of Imaging, Diagnostics, and Software Reliability. It builds anatomically inspired multi-layer, multi-component 3D skin models with a growing lesion and digitally renders them, giving synthetic skin images with pixel-level lesion segmentation masks under controlled variation of skin colour (melanosome fraction), hair, blood fraction, lesion shape, lesion material, lighting and rotation. The authors show that comparative AI performance trends on S-SYNTH follow those on real dermatologic images, so it can be used to study bias across skin tones and to augment small patient datasets. Folders encode the generating parameters (skin model, hair model, melanosome and blood fractions, lesion, timepoint, light), and each leaf holds image.png, mask.png, cropped versions and crop_size.txt. As measured 2026-11-14 the repository uses about 327 GB under data/, with dataset_splits and outputs folders; load with the datasets library (load_dataset('didsr/ssynth_data')) or read PNGs directly. The FDA authors release it under CC0 1.0; cite Kim et al. 2024 (arXiv:2408.00191).","displayTitle":"S-SYNTH: synthetic dermoscopy-style skin lesion images with segmentation masks (FDA)","id":"5bddb473-a761-5cf3-b406-a49ab7e151d0","license":"CC0-1.0","licenseEvidenceUrl":"https://huggingface.co/datasets/didsr/ssynth_data","objectCountEstimate":1029,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2408.00191","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"didsr/ssynth_data\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"didsr/ssynth_data\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=didsr, license:cc0-1.0, size_categories:1K<n<10K, downloads=1321"],"format":"mixed","generatedAt":"2026-11-14T14:15:18.862Z","layout":{"example":"data/outputs/segmentation_results/artifacts/MODEL-dermosegdiff--all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--00--dsd_i01--n-dsd_i01_s-128_b-32_t-250_sc-linear_best.pth-DATASET-all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--test_images.csv","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"individual_image_name,tone,ita,model_name,dice,hausdroff,mean_squared_diameter\n/projects01/VICTRE/elena.sizikova/skin/real_dataset/ISIC2018/ISIC2018_Task1-2_Training_Input/ISIC_0011329.jpg,lt1,43.9244060541768,/projects01/VICTRE/elena.sizikova/ssynth_artifact/models_step/dermosegdiff/all_tones_real_ISIC_1.0_add_synth_0.0_None_10k/00/dsd_i01/n-dsd_i01_s-128_b-32_t-250_sc-linear_best.pth,0.8814582824707031,17.804493814764857,0.0\n/projects01/VICTRE/elena.sizikova/skin/real_dataset/ISIC2018/ISIC2018_Task1-2_Tra","encoding":"utf8","key":"data/outputs/segmentation_results/artifacts/MODEL-dermosegdiff--all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--00--dsd_i01--n-dsd_i01_s-128_b-32_t-250_sc-linear_best.pth-DATASET-all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--test_images.csv","truncatedAt":512}],"sampleObjects":["data/dataset_splits/all_tones_real_HAM_0.2/test_images.txt","data/dataset_splits/all_tones_real_HAM_0.2/test_masks.txt","data/dataset_splits/all_tones_real_HAM_0.2/train_images.txt","data/dataset_splits/all_tones_real_HAM_0.2/train_masks.txt","data/dataset_splits/all_tones_real_HAM_0.2/val_images.txt"],"sidecars":[{"bytes":"---\nlicense: cc0-1.0\ntask_categories:\n- image-segmentation\ntags:\n- medical\npretty_name: S-SYNTH\nsize_categories:\n- 1K<n<10K\n---\n# S-SYNTH\n<!-- Provide a quick summary of the dataset. -->\nS-SYNTH is an open-source, flexible skin simulation framework to rapidly generate synthetic skin models and images using digital rendering of an anatomically inspired multi-layer, multi-component skin and growing lesion model. It allows for generation of highly-detailed 3D skin models and digitally rendered synthetic images of diverse human skin tones, with full control of underlying parameters and the image formation process.\n\n## Framework Details\nS-SYNTH can be used to generate synthetic skin images with annotations (including segmentation masks) with variations of:\n  * Hair artifact\n  * Blood fraction\n  * Lesion shape\n  * Melanosome fraction\n  * Sample per pixel (for rendering)\n  * Lighting condition\n\n<p align=\"center\">\n<img src=\"https://raw.githubusercontent.com/DIDSR/ssynth-release/main/images/variation.png\" width=\"700\">\n</p>\n\n### Framework Description\n\n<!-- Provide a longer summary of what this dataset is. -->\n\n- **Curated by:** [Andrea Kim](https://www.linkedin.com/in/andreakim91), [Niloufar Saharkhiz](https://www.linkedin.com/in/niloufar-saharkhiz/), [Elena Sizikova](https://esizikova.github.io/), [Miguel Lago](https://www.linkedin.com/in/milaan/), [Berkman Sahiner](https://www.linkedin.com/in/berkman-sahiner-6aa9a919/), [Jana Delfino](https://www.linkedin.com/in/janadelfino/), [Aldo Badano](https://www.linkedin.com/in/aldobadano/)\n- **License:** Creative Commons 1.0 Universal License (CC0)\n\n### Dataset Sources\n\n<!-- Provide the basic links for the dataset. -->\n\n- **Paper:** [https://arxiv.org/abs/2408.00191](https://arxiv.org/abs/2408.00191)\n- **Code:** [https://github.com/DIDSR/ssynth-release](https://github.com/DIDSR/ssynth-release)\n- **Data:** [https://huggingface.co/datasets/didsr/ssynth_data](https://huggingface.co/datasets/didsr/ssynth_data)\n\n## Uses\n\n<!-- Address questions around how the dataset is intended to be used. -->\n\nS-SYNTH is intended to facilitate augment limited patient datasets and identify AI performance trends by generating synthetic skin data.\n\n### Direct Use\n\n<!-- This section describes suitable use cases for the dataset. -->\n\nS-SYNTH can be used to evaluate comparative AI performance trends in dermatologic analysis tasks, estimating the effect skin appearance, such as skin color, presence of hair, lesion shape, and blood fraction on AI performance in lesion segmentation. \nS-SYNTH can be used to mitigate effects of mislabeled examples and limitations of a limited patient dataset.\nS-SYNTH can be used to either train or test pre-trained AI models.\n\n### Out-of-Scope Use\n\n<!-- This section addresses misuse, malicious use, and uses that the dataset will not work well for. -->\n\nS-SYNTH cannot be used in lieu of real patient examples to make performance determinations.\n\n## Dataset Structure\n\n<!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->\n\nS-SYNTH is organized into a directory structure that indicates the parameters. The folder\n```\ndata/synthetic_dataset/output_10k/output/skin_[SKIN_MODEL_ID]/hairModel_[HAIR_MODEL_ID]/mel_[MELANOSOME_FRACTION]/fB_[BLOOD_FRACTION]/lesion_[LESION_ID]/T_[TIMEPOINT]/[LESION_MATERIAL]/hairAlb_[HAIR_ALBEDO]/light_[LIGHT_NAME]/ROT[LESION_ROTATION_VECTOR]_[RANDOM_ANGLE]/\n```\ncontains image files and paired segmentation masks with the specified parameters.\n\n```\n$ tree data/synthetic_dataset/output_10k/output/skin_099/hairModel_098/mel_0.47/fB_0.05/lesion_9/T_050/HbO2x1.0Epix0.25/hairAlb_0.84-0.6328-0.44/light_hospital_room_4k/ROT0-0-1_270/\n\ndata/synthetic_dataset/output_10k/output/skin_099/hairModel_098/mel_0.47/fB_0.05/lesion_9/T_050/HbO2x1.0Epix0.25/hairAlb_0.84-0.6328-0.44/light_hospital_room_4k/ROT0-0-1_270/\n├── mask.png\n├── image.png\n├── cropped_mask.png\n├── cropped_mask.png\n└── crop_size.txt\n```\nE","encoding":"utf8","key":"README.md","name":"README.md","size":6211,"truncated":true},{"bytes":"{\n  \"builder_name\": \"generator\",\n  \"citation\": \"\",\n  \"config_name\": \"default\",\n  \"dataset_name\": \"generator\",\n  \"dataset_size\": 5170792,\n  \"description\": \"\",\n  \"download_checksums\": {},\n  \"download_size\": 0,\n  \"features\": {\n    \"input_image\": {\n      \"_type\": \"Image\"\n    },\n    \"edit_prompt\": {\n      \"dtype\": \"string\",\n      \"_type\": \"Value\"\n    },\n    \"edited_image\": {\n      \"_type\": \"Image\"\n    }\n  },\n  \"homepage\": \"\",\n  \"license\": \"\",\n  \"size_in_bytes\": 5170792,\n  \"splits\": {\n    \"train\": {\n      \"name\": \"train\",\n      \"num_bytes\": 5170792,\n      \"num_examples\": 8000,\n      \"dataset_name\": \"generator\"\n    }\n  },\n  \"version\": {\n    \"version_str\": \"0.0.0\",\n    \"major\": 0,\n    \"minor\": 0,\n    \"patch\": 0\n  }\n}","encoding":"utf8","key":"data/synthetic_dataset/hf_datasets/bloodVessel_remove_ssynth_train/train/dataset_info.json","name":"dataset_info.json","size":718,"truncated":false},{"bytes":"{\n  \"builder_name\": \"generator\",\n  \"citation\": \"\",\n  \"config_name\": \"default\",\n  \"dataset_name\": \"generator\",\n  \"dataset_size\": 5168074,\n  \"description\": \"\",\n  \"download_checksums\": {},\n  \"download_size\": 0,\n  \"features\": {\n    \"input_image\": {\n      \"_type\": \"Image\"\n    },\n    \"edit_prompt\": {\n      \"dtype\": \"string\",\n      \"_type\": \"Value\"\n    },\n    \"edited_image\": {\n      \"_type\": \"Image\"\n    }\n  },\n  \"homepage\": \"\",\n  \"license\": \"\",\n  \"size_in_bytes\": 5168074,\n  \"splits\": {\n    \"train\": {\n      \"name\": \"train\",\n      \"num_bytes\": 5168074,\n      \"num_examples\": 8000,\n      \"dataset_name\": \"generator\"\n    }\n  },\n  \"version\": {\n    \"version_str\": \"0.0.0\",\n    \"major\": 0,\n    \"minor\": 0,\n    \"patch\": 0\n  }\n}","encoding":"utf8","key":"data/synthetic_dataset/hf_datasets/bloodVessel_verb_ssynth_train/train/dataset_info.json","name":"dataset_info.json","size":718,"truncated":false},{"bytes":"{\n  \"builder_name\": \"generator\",\n  \"citation\": \"\",\n  \"config_name\": \"default\",\n  \"dataset_name\": \"generator\",\n  \"dataset_size\": 2009470,\n  \"description\": \"\",\n  \"download_checksums\": {},\n  \"download_size\": 0,\n  \"features\": {\n    \"input_image\": {\n      \"_type\": \"Image\"\n    },\n    \"edit_prompt\": {\n      \"dtype\": \"string\",\n      \"_type\": \"Value\"\n    },\n    \"edited_image\": {\n      \"_type\": \"Image\"\n    }\n  },\n  \"homepage\": \"\",\n  \"license\": \"\",\n  \"size_in_bytes\": 2009470,\n  \"splits\": {\n    \"train\": {\n      \"name\": \"train\",\n      \"num_bytes\": 2009470,\n      \"num_examples\": 3174,\n      \"dataset_name\": \"generator\"\n    }\n  },\n  \"version\": {\n    \"version_str\": \"0.0.0\",\n    \"major\": 0,\n    \"minor\": 0,\n    \"patch\": 0\n  }\n}","encoding":"utf8","key":"data/synthetic_dataset/hf_datasets/calChart_remove_ssynth_train/train/dataset_info.json","name":"dataset_info.json","size":718,"truncated":false}],"sourceObjectKey":"data/outputs/segmentation_results/artifacts/MODEL-dermosegdiff--all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--00--dsd_i01--n-dsd_i01_s-128_b-32_t-250_sc-linear_best.pth-DATASET-all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--test_images.csv","starterCell":"import pandas as pd\ndf = pd.read_csv(\"hf://datasets/didsr/ssynth_data/data/outputs/segmentation_results/artifacts/MODEL-dermosegdiff--all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--00--dsd_i01--n-dsd_i01_s-128_b-32_t-250_sc-linear_best.pth-DATASET-all_tones_real_ISIC_1.0_add_synth_0.0_None_10k--test_images.csv\")\ndf.head()","starterSamplePath":"data/dataset_splits/all_tones_real_HAM_0.2/test_images.txt","totalObjects":1029},"summary":"S-SYNTH is a knowledge-based, open-source skin simulation framework and pre-generated dataset from the US Food and Drug Administration's Division of Imaging, Diagnostics, and…","title":"S-SYNTH: synthetic dermoscopy-style skin lesion images with segmentation masks (FDA)","updatedAt":"2026-11-14T17:42:28.377Z","visibility":"public","websiteUrl":"https://github.com/DIDSR/ssynth-release"}}