SkinDisNet: A Multi-Class Clinical Images and Metadata for Skin Disease

Published: 26 June 2025| Version 2 | DOI: 10.17632/yj3md44hxg.2
Contributors:
,
,
,
,

Description

Purpose : Classification and identification of six different skin disease categories for automatic diagnosis. Type of data: Image files (512 x 512 pixels) Data format: Joint Photographic Expert Group (JPG) and Comma Separated Values (CSV) file formats Number of classes: Six (Atopic Dermatitis, Contact Dermatitis, Eczema, Scabies, Seborrheic Dermatitis, and Tinea Corporis) Number of images: Preprocessed Folder: 1710 images Augmented Folder: 11970 images Metadata: The metadata associated with each skin lesion is composed of 7 attributes. All attributes are available in a CSV document. In total, there are 416 patients and 1,710 skin disease images present in the dataset. Each image/sample has a reference to the patient and the skin disease in the metadata. Data Acquisition: Images were captured using smartphone cameras during the patients’ consultations with dermatologists. Data source: Clinical sources: 1. Institution: Rangpur Medical College, City: Rangpur, Country: Bangladesh 2. Institution: Shahid Syed Nazrul Islam Medical College, City: Kishoreganj, Country: Bangladesh Applications: Skin disease detection and classification, diagnosis systems, medical image analysis, computer vision and more.

Files

Institutions

  • Begum Rokeya University

Categories

Computer Vision, Dermatology, Machine Learning, Image Classification, Medical Image Processing

Funders

  • ICT Division, Government of the People’s Republic of Bangladesh
    Grant ID: 1280101-120008431-3631108

Licence