Sign in to view source links and access this dataset
Description
PAD-UFES-20 contains 2,298 smartphone clinical images of skin lesions from 1,373 patients, covering 1,641 distinct lesions. The dataset includes tabular metadata and is mirrored on Hugging Face by SalmaneExploring for reproducible teledermatology experiments. Labels include actinic keratosis, basal cell carcinoma, melanoma, nevus, squamous cell carcinoma, and seborrheic keratosis.
Use Cases
Training skin lesion classification models based on labeled smartphone images.
Developing multimodal diagnostic tools that combine image and clinical metadata.
Benchmarking teledermatology algorithms for reproducibility studies.
Studying the visual characteristics of specific lesion types like melanoma or basal cell carcinoma.
Strengths
Contains 2,298 clinical images from 1,373 patients, providing a patient-level perspective.
Covers 1,641 distinct skin lesions with diagnostic labels for six conditions.
Includes both image data and tabular clinical metadata, enabling multimodal analysis.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count for the tabular metadata is unknown, which may limit suitability assessment.
Freshness should be verified as the last update timestamp is in the future (2026-04-28).
Provenance
Source
Mirrored from the original PAD-UFES-20 dataset.
Collection Method
Likely collected via smartphone imaging in a clinical setting.
Freshness
Last updated 2026-04-28 10:22:16; freshness should be verified.
License is unknown; users should verify terms of use before application.