Skip to content

sfu-mial/awesome-skin-image-analysis-datasets

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

Datasets for skin image analysis research

Awesome

This is a collection of datasets used for skin image analysis research. More details about some of these datasets can be found in our surveys:

  • J. Kawahara, G. Hamarneh, "Visual Diagnosis of Dermatological Disorders: Human and Machine Performance", arXiv pre-print arXiv:1906.01256, 2019. [DOI]
  • Z. Mirikharaji, K. Abhishek, A. Bissoto, C. Barata, S. Avila, E. Valle, M. E. Celebi, G. Hamarneh, "A Survey on Deep Learning for Skin Lesion Segmentation", Medical Image Analysis: 102863, 2023. [DOI]

To list your or any other dataset(s) here, please create an issue.

Dataset Link Notes
ACNE04 https://github.com/xpwu95/LDL
BCN_20000 https://challenge2019.isic-archive.com/data.html Part of ISIC 2019
Asan and Hallym https://figshare.com/articles/Asan_and_Hallym_Dataset_Thumbnails_/5406136
Atlas Dermatologico http://www.atlasdermatologico.com.br
Derm12345 https://arxiv.org/abs/2406.07426
DermQuest http://dermquest.com
DermAtlas http://www.dermatlas.net
Dermis http://www.dermis.net/dermisroot/en/home/index.htm
Dermnet http://www.dermnet.com
DermWeb http://www.dermweb.com
Dermnet NZ https://www.dermnetnz.org
Dermatoweb http://www.dermatoweb.net
Danderm http://www.danderm-pdv.is.kkh.dk/atlas/index.html
Dermatologia Praktyczna http://derma.pl/
DermSynth3D https://github.com/sfu-mial/DermSynth3D
DermX https://github.com/ralucaj/dermx 525 dermatological images with diagnoses and diagnosis explanations by three dermatologists
Dermofit https://licensing.edinburgh-innovations.ed.ac.uk/i/software/dermofit-image-library.html
Derm7pt http://derm.cs.sfu.ca/
Diverse Dermatology Images (DDI) https://ddi-dataset.github.io
ENHANCE (ENriching Health data by ANnotations of Crowd and Experts) https://github.com/raumannsr/ENHANCE ABC criteria annotations of ISIC 2017 and PH2 datasets
Fitzpatrick17k https://github.com/mattgroh/fitzpatrick17k 16,577 clinical images with diagnosis and Fitzpatrick scale labels
HAM10000 https://www.nature.com/articles/sdata2018161
Hellenic Derm Atlas http://www.hellenicdermatlas.com/en
ISIC Archive1 https://isic-archive.com/ See below
Monkeypox-dataset-2022 [Ahsan et al.] https://github.com/mahsan2/Monkeypox-dataset-2022
Monkeypox Skin Image Dataset 2022 [Islam et al.] https://www.kaggle.com/datasets/arafathussain/monkeypox-skin-image-dataset-2022
MedMNIST https://medmnist.com DermaMNIST. a subset of MedMNIST, contains skin images
Med-Node http://www.cs.rug.nl/~imaging/databases/melanoma_naevi/
Meddean http://www.meddean.luc.edu/lumen/MedEd/medicine/dermatology/melton/atlas.htm
MoleMap https://molemap.co.nz
MSK https://arxiv.org/abs/1710.05006
PH2 https://www.fc.up.pt/addi/ph2%20database.html
PAD-UFES-20 https://data.mendeley.com/datasets/zr7vgbcyr2/1 clinical skin lesion images from smartphones
Skin3D https://github.com/jeremykawahara/skin3d
SD198 https://drive.google.com/file/d/1YgnKz3hnzD3umEYHAgd29n2AwedV1Jmg/view
Skin Cancer Detection https://uwaterloo.ca/vision-image-processing-lab/research-demos/skin-cancer-detection
SKINCON https://skincon-dataset.github.io/index.html
SymDerm http://opendemo.uib.es/datasets/symderm-dataset/ expert annotation of lesion symmetry across 2 perpendicular axes
University of Iowa Clinical Skin Disease Images http://www.medicine.uiowa.edu/dermatology/diseaseimages/
UWaterloo Skin Cancer Detection dataset https://uwaterloo.ca/vision-image-processing-lab/research-demos/skin-cancer-detection images taken from DermIS and DermQuest along with lesion segmentation
XiangyaDerm http://airl.csu.edu.cn/xiangyaderm/

1 The Gallery section of the ISIC Archive includes the following datasets:

  • 2018 JID Editorial Images (100 images)
  • Dermoscopedia (CC-BY) (5 images)
  • HAM10000 (10015 images)
  • MSK-1 (1100 images), MSK-2 (1535 images), MSK-3 (225 images), MSK-4 (947 images), MSK-5 (111 images)
  • SONIC (9251 images)
  • UDA-1 (557 images), UDA-2 (60 images)