Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?


Failed to load latest commit information.




We annotated 16,577 clinical images sourced from two dermatology atlases — DermaAmin and Atlas Dermatologico — with Fitzpatrick skin type labels with two data annotation services: Scale AI and Centaur Labs. The Fitzpatrick labeling system, while not perfect, is a six-point scale originally developed for classifying sun reactivity of skin phenotype. The Fitzpatrick scale served as the basis for skin color in emojis and, more recently, the Fitzpatrick scale has been used in computer vision applications to evaluate algorithmic fairness and model accuracy. The annotated images represent 114 skin conditions with at least 53 images and a maximum of 653 images per skin condition.


We've included the original image sources, the training script, and a notebook to compare Fitzpatrick annotations with individual typology angle scores. You can read the dataset and our analysis in our paper.

We thank Scale AI and Centaur Labs for providing Fitzpatrick Skin Type annotations for all images in this dataset for free.

Updates July 7 2022

Given the subjectivity of annotating Fitzpatrick Skin Type in images showing skin disease, we evaluated how well experts, crowds, and an algorithm compare with respect to inter-rater reliability. You can find replication files for this analysis in the annotation_evaluation folder.

Data Usage

Download the dataset

You can find the Fitzpatrick annotations in fitzpatrick17k.csv. You can download the images from their original source, which is shared in the url column of the Fitzpatrick annotations .csv. Alternatively, fill out this form and contact us and we can provide a link to all the images.

Replicate our analysis

The results from our paper can be replicated using and ita_fitzpatrick_analysis.ipynb.

After you download the dataset, edit by specifying the image directory of the dataset, and then run python 20 full where 20 refers to the number of epochs and full refers to the full dataset.

You can check out our comparison of Fitzpatrick annotations and individual typology angle scores with the ita_fitzpatrick_analysis.ipynb

How to cite this dataset and paper

  title={Evaluating deep neural networks trained on clinical images in dermatology with the fitzpatrick 17k dataset},
  author={Groh, Matthew and Harris, Caleb and Soenksen, Luis and Lau, Felix and Han, Rachel and Kim, Aerin and Koochek, Arash and Badri, Omar},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},

And the second paper:

  title={Towards transparency in dermatology image datasets with skin tone annotations by experts, crowds, and an algorithm},
  author={Groh, Matthew and Harris, Caleb and Daneshjou, Roxana and Badri, Omar and Koochek, Arash},
  journal={arXiv preprint arXiv:2207.02942},


Original images collected from Atlas Dermatologico and DermaAmin

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.