The BCN20000 dataset consists of dermoscopic images of skin lesions taken between 2010 and 2016 at the Hospital Clínic in Barcelona. It aims to tackle the challenge of classifying dermoscopic images of skin cancer under diverse conditions, including lesions in challenging locations (such as nails and mucosa), large lesions exceeding the dermoscopy device's aperture, and hypo-pigmented lesions. A detailed data descriptor is available: Hernández-Pérez, C., Combalia, M., Podlipnik, S. et al. BCN20000: Dermoscopic Lesions in the Wild. Sci Data 11, 641 (2024). https://doi.org/10.1038/s41597-024-03387-w

Test set from the ISIC 2016 Challenge. The "Skin Lesion Analysis Towards Melanoma Detection" challenge leverages a dataset of annotated skin lesion images from the ISIC Archive, The dataset contains a representative mix of images of both malignant and benign skin lesions.

Test set from the ISIC 2017 Challenge.

Validation set from the ISIC 2017 Challenge.

Training set from the ISIC 2018 Challenge. When using the ISIC 2018 datasets in your research, please cite the following works: > [1] Noel Codella, Veronica Rotemberg, Philipp Tschandl, M. Emre Celebi, Stephen Dusza, David Gutman, Brian Helba, Aadi Kalloo, Konstantinos Liopyris, Michael Marchetti, Harald Kittler, Allan Halpern: "Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)", 2018; arxiv.org/abs/1902.03368 > > [2] Tschandl, P., Rosendahl, C. & Kittler, H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci. Data 5, 180161 doi:10.1038/sdata.2018.161 (2018).

Test set from the ISIC 2018 Challenge. When using the ISIC 2018 datasets in your research, please cite the following works: > [1] Noel Codella, Veronica Rotemberg, Philipp Tschandl, M. Emre Celebi, Stephen Dusza, David Gutman, Brian Helba, Aadi Kalloo, Konstantinos Liopyris, Michael Marchetti, Harald Kittler, Allan Halpern: "Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)", 2018; arxiv.org/abs/1902.03368 > > [2] Tschandl, P., Rosendahl, C. & Kittler, H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci. Data 5, 180161 doi:10.1038/sdata.2018.161 (2018).

Validation set from the ISIC 2018 Challenge. When using the ISIC 2018 datasets in your research, please cite the following works: > [1] Noel Codella, Veronica Rotemberg, Philipp Tschandl, M. Emre Celebi, Stephen Dusza, David Gutman, Brian Helba, Aadi Kalloo, Konstantinos Liopyris, Michael Marchetti, Harald Kittler, Allan Halpern: "Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)", 2018; arxiv.org/abs/1902.03368 > > [2] Tschandl, P., Rosendahl, C. & Kittler, H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci. Data 5, 180161 doi:10.1038/sdata.2018.161 (2018).

Training set from the ISIC 2019 Challenge.

A small study evaluated the accuracy of smartphone and web-based dermatology apps offering AI diagnostics using an independent test set of clinical images. Sun MD, Kentley J, Mehta P, Dusza S, Halpern AC, Rotemberg V. Accuracy of commercially available smartphone applications for the detection of melanoma. Br J Dermatol. 2022;186(4):744-746. doi:10.1111/bjd.20903

A collection of 248 melanocytic lesions that were submitted by experts as exemplars for 1 out of 31 dermoscopic features (8 images per feature), and used for evaluating agreement among (unrelated) experts on (1) malignancy, (2) feature presence, and (3) feature localization within a lesion. The repository of image masks and superpixel annotations is here: https://github.com/ISIC-Research/expert-annotation-agreement-data

A dataset of clinical and dermoscopy images of skin lesions collected in Argentina as a reference for the evaluation of AI tools in this population

A dataset of clinical and dermoscopy images of skin lesions collected in Argentina as a reference for the evaluation of AI tools in this population. Supplementary data and an exploratory analysis of the data are publicly available at https://github.com/piashiba/HIBASkinLesionsDataset.

These are the images used in the paper: Analysis of the ISIC image datasets: Usage, benchmarks and recommendations Paper Link: https://www.sciencedirect.com/science/article/pii/S1361841521003509 They have also been used by newer versions such as: Skin Lesion Classification Using Dermoscopic Images and Clinical Metadata: Insights from Multimodal Models Paper Link: https://openaccess.thecvf.com/content/CVPR2025W/MULA2025/papers/Ahammed_Skin_Lesion_Classification_Using_Dermoscopic_Images_and_Clinical_Metadata_Insights_CVPRW_2025_paper.pdf https://api.isic-archive.com/collections/469/

## The iToBoS dataset: skin region images extracted from 3D total body photographs for lesion detection The early detection of skin cancer is critical for improving patient outcomes. Traditionally, dermatologists rely on dermoscopy to examine pigmented skin lesions. While this non-invasive technique enhances diagnostic accuracy, its effectiveness is highly dependent on the clinician’s expertise. Additionally, capturing dermoscopic images for every suspicious lesion is a labor-intensive process. Given these challenges, there is an increasing need for computer-aided diagnosis (CAD) systems that utilize conventional cameras. Such systems can support general physicians and other non-specialist practitioners in identifying potential malignant lesion, improving early detection and intervention. Moreover, they facilitate longitudinal tracking of lesions, aiding researchers in studying disease progression and treatment efficacy. This dataset provides high-resolution skin patch images extracted from 3D total body photographs to support the development of advanced machine learning models for lesion detection. It serves as a valuable resource for researchers working on automated skin lesion analysis, particularly in the context of total body photography (TBP). This dataset contains 59,997 lesion identifying regions-of-interest (ROIs) embedded in 16,954 images stemming from 100 patients. ## Dataset Description: The iToBoS dataset consists of 16,954 high-resolution images of skin regions obtained from anonymized 3D avatars of patients. These avatars were generated using the Canfield VECTRA WB360 system, a cutting-edge imaging technology that captures comprehensive, full-body skin images using 92 fixed cameras arranged in 46 stereo pairs with xenon flash lighting. The images were collected from patients at two clinical sites: the Clinical Hospital of Barcelona (Spain) (n=7,729) and the University of Queensland (Australia) (n=9,225). The dataset provides diverse anatomical locations, including the torso, arms, and legs, with each image having an average resolution of 1012x827 pixels and a 45-pixel overlap between adjacent images. The images are extracted from 3D avatars while ensuring compliance with GDPR regulations by automatically removing patient facial features. Each image is accompanied by metadata, including patient age range, body location, and sun damage score, allowing for in-depth analysis and stratification. ## Significance of the Dataset: 1. **Facilitates Automated Skin Lesion Detection:** The dataset supports the development of AI-based lesion detection models that can improve early diagnosis of skin cancer, particularly in regions with limited access to dermatological expertise. 2. **Supports Total Body Photography Research:** Leveraging 3D TBP for lesion detection is an emerging field, and this dataset provides a benchmark for further exploration. 3. **Enhances Machine Learning Applications:** The dataset serves as a benchmark for developing state-of-the-art computer vision and deep learning models for detection of skin lesions. ## PNG Format The image files were originally captured in PNG format, but are published here in compressed JPEG format. Our internal testing indicates that over 97% of the JPEG images achieve a PSNR greater than 35dB when compared to the original PNG versions, while being only ~6% of the original dataset size. Additionally, the original PNG files are available [on Figshare](https://figshare.com/articles/dataset/iToBoS_2024_-_Skin_Lesion_Detection_with_3D-TBP/28452545). ## Funding EC | EU Framework Programme for Research and Innovation H2020 | H2020 Priority Societal Challenges | H2020 Health (H2020 Societal Challenges - Health, Demographic Change and Well-being) - SC1-BHC-06-2020-965221

Digital camera photographs of the posterior trunk of patients at risk for melanoma, taken at UPMC Hillman Cancer Center as part of the 96-099 image/tissue banking protocol. Images from "Development and narrow validation of computer vision approach to facilitate assessment of change in pigmented cutaneous lesions," corresponding author John M Kirkwood.

Collection defined by Dr. Jorge A. Rios-Duarte for a research project.

*MILK10k* consists of 10480 images, each representing a paired clinical close-up and dermatoscopic image for 5240 lesions. The dataset’s metadata include age (in 5-year intervals), sex, anatomic site, skin tone, diagnosis, method of ground truth establishment (histopathology or other means), and, if a dermatoscopic image of the same lesion was previously included in ISIC, its corresponding ISIC identifier. Skin tone is categorized into six levels, ranging from very dark (0) to very light (5), intentionally distinct from the Fitzpatrick skin types to avoid confusion. Most patients had skin tones in the middle ranges. Of the 5240 lesions, 95.7% were biopsied or excised, with histopathology serving as the gold standard for diagnosis. Diagnoses were mapped to both the ISIC-Dx diagnostic scheme and a simplified classification based on the ISIC2018/2019 challenge and HAM10000 diagnostic categories. The dataset includes 11 broad diagnostic categories: 1. Basal cell carcinoma (bcc) 2. Melanocytic nevus (nv) 3. Benign keratinocytic lesion (bkl) 4. Squamous cell carcinoma/keratoacanthoma (sccka) 5. Melanoma (mel) 6. Actinic keratosis/intraepidermal carcinoma (akiec) 7. Dermatofibroma (df) 8. Inflammatory and infectious conditions (inf) 9. Vascular lesions and hemorrhage (vasc) 10. Other benign proliferations including collision tumors (ben_oth) 11. Other malignant proliferations including collision tumors (mal_oth) Additionally, we provide the most specific ISIC-Dx diagnosis and its parent branch in the ISIC-Dx diagnostic tree. In cases where a dermatoscopic image of the same lesion was already included in the ISIC archive, its ISIC identifier is reported in the metadata. Furthermore, all images have been annotated using the MONET framework, with probabilities for the following concept term groups included in the metadata: 1. Ulceration, crust 2. Hair 3. Vasculature, vessels 4. Erythema 5. Pigmentation 6. Gel, water drop, fluid, dermoscopy liquid 7. Skin markings, pen ink, purple pen In addition to *MILK10k*, we have curated a smaller benchmark dataset, called *MILK10k Benchmark* derived from the same sources and covering the same diagnostic categories. This dataset is available as part of a live challenge within the ISIC framework and can be accessed on ISIC. Images were provided by the following institutions: - Department of Dermatology, Medical University of Vienna, Vienna, Austria - Medicine Faculty Department of Dermatology, Ankara University, Ankara, Turkey - Mayne Academy of General Practice, Medical School, The University of Queensland, Australia - Dermatology Service, Memorial Sloan Kettering Cancer Center, New York, USA - Independent Researcher, 1000 Skopje, North Macedonia

Both benign and malignant melanocytic lesions. Almost all diagnoses were confirmed by histopathology reports; the remainder consists of benign lesions confirmed by clinical follow-up. Images were not taken with modern digital cameras.

Assorted images and lesions, mostly nevi and basal cell carcinomas. These images were found based on a search not filtered for any particular pathology. All diagnoses confirmed by histopathology.

Seborrheic keratoses obtained from patients during a clinical visit. These lesions were not biopsied and were determined to be seborrheic keratoses by agreement of three experts.

This dataset contains detailed skin tone annotations collected from a prospective, single-center observational study performed at Memorial Sloan Kettering Cancer Center from 2023-2024. The cohort consists of 64 adult patients who underwent full-body skin examinations by board-certified dermatologists. To ensure diverse representation across the spectrum of skin tones, patients were recruited to achieve a balanced distribution across all six Fitzpatrick Skin Types. This dataset was developed to evaluate the reliability of different skin tone labeling methods and to support fairness research in dermatologic AI. The dataset comprises both patient-level and site-level metadata for skin tone classification using the Fitzpatrick Skin Type scale, Monk Skin Tone scale, Pantone SkinTone Guide, and colorimeter readings (SkinColorCatch, Delfin Technologies). A total of 4,879 dermoscopic images are included. Skin tone assessments were collected across both lesional and non-lesional (normal skin) sites, mapped to standardized anatomic locations. All skin lesions are assumed to be benign, as they were imaged immediately following dermatologic evaluation. All data were collected under an IRB-approved protocol with informed consent. The dataset has been fully de-identified in accordance with HIPAA regulations, and no protected health information (PHI) is included.

A collection of dermoscopic images of biopsied melanocytic lesions. This was used to test for performance degradation of an AI melanoma classifier in discriminating melanomas from nevi among newly-acquired melanocytic lesions versus longer-existing acquired melanocytic lesions.

We conducted a prospective, observational clinical validation study to assess the diagnostic accuracy of the AI algorithm (ADAE) in predicting melanoma from dermoscopy skin lesion images. Patients who had consented for a skin biopsy to exclude melanoma were eligible. All lesions underwent biopsy.