DermaMNIST

Other names: MedMNIST v2 DermaMNIST, dermamnist

Overview

The DermaMNIST is based on the HAM10000, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. The dataset consists of 10,015 dermatoscopic images categorized as 7 different diseases, formulized as a multi-class classification task.

Description (from the source)

The DermaMNIST is based on the HAM10000, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. The dataset consists of 10,015 dermatoscopic images categorized as 7 different diseases, formulized as a multi-class classification task. We split the images into training, validation and test set with a ratio of 7:1:2. The source images of 3×600×450 are resized into 3×28×28.

Domain
Medical & Healthcare
Data type
Images
Tasks
Medical Image Classification
Subfields
Dermatology, Medical image classification
Languages
Not available
Tags
2d

Statistics and structure

Number of samples
10,015
Number of classes
7
Number of features
Not available
Number of subjects / cases
Not available
Number of images
10,015
Number of files
Not available
Size
Not available
Formats
NPZ (NumPy)
Missing values
Not available

Samples note: train 7,007 + val 1,003 + test 2,005.

Licence and access

Licence
Creative Commons Attribution-NonCommercial 4.0 (CC-BY-NC-4.0)
Licence documentation
Stated by the publisher
Commercial use
No
Redistribution
Yes
Attribution required
Yes
Access method
Direct download
Access restrictions
Not available
Terms of use
Not available

Allows non-commercial use only, with attribution. Licence text

Always review the original publisher's terms before use; the explorer grants no licence.

Source and download

Platform
Zenodo
Publisher / source
MedMNIST (official package)
Identifier at the source
medmnist:dermamnist
DOI
10.5281/zenodo.10519652
Metadata source
medmnist
Metadata last updated
2026-11-14

The explorer does not host data files; all links point to the original publisher.

Related papers

  1. Paper that introduced the dataset
    Jiancheng Yang, Rui Shi, Donglai Wei, Zequan Liu, Lin Zhao, Bilian Ke, Hanspeter Pfister, Bingbing Ni (2023). MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification. Scientific Data. https://doi.org/10.1038/s41597-022-01721-8
    Listed for this dataset in the source catalogue · Paper link · Evidence

Cite the dataset

Built from recorded fields only; missing fields are omitted, not guessed. It is also good practice to cite the paper that introduced the dataset, if there is one.

APA 7
DermaMNIST [Data set]. (2023). MedMNIST (official package). https://doi.org/10.5281/zenodo.10519652
IEEE
[1] "DermaMNIST," MedMNIST (official package), 2023. [Online]. Available: https://doi.org/10.5281/zenodo.10519652 (DOI: 10.5281/zenodo.10519652)
BibTeX
@misc{dermamnist2023,
  title = {{DermaMNIST}},
  year = {2023},
  publisher = {MedMNIST (official package)},
  doi = {10.5281/zenodo.10519652},
  url = {https://medmnist.com/},
  howpublished = {\url{https://doi.org/10.5281/zenodo.10519652}},
  note = {Dataset. Accessed: 2026-11-21}
}

Verification

Verification status
Verified
Last link check
2026-11-15
Source verification date
2026-11-15
Verification method
Automated check: HTTP status of the original link, the name present on the page, a licence mention, download and documentation links, DOI check.
Notes
Sample counts, labels, licence and Zenodo file URL are stated by the dataset authors. MedMNIST images are pre-processed (resized) versions of their source datasets; the source datasets keep their own terms. Original URL not yet checked for availability.
Source and check date of each field
FieldSourceDate
descriptionmedmnist 3.0.2 package, medmnist/info.py (INFO["dermamnist"])2026-11-14
number_of_samplesmedmnist 3.0.2 package, medmnist/info.py (INFO["dermamnist"])2026-11-14
number_of_classesmedmnist 3.0.2 package, medmnist/info.py (INFO["dermamnist"])2026-11-14
download_urlmedmnist 3.0.2 package, medmnist/info.py (INFO["dermamnist"])2026-11-14
doimedmnist 3.0.2 package README (Data: Zenodo DOI)2026-11-14
licensemedmnist 3.0.2 package, medmnist/info.py (INFO["dermamnist"])2026-11-14
Check history (7)
CheckOutcomeHTTPDate
DOIAvailable2002026-11-15
Repository urlAvailable2002026-11-15
Documentation linkAvailable2002026-11-15
Download linkAvailable2002026-11-15
LicenceLicence matches—2026-11-15
Name on the pageName matches—2026-11-15
Original linkAvailable2062026-11-15

Similar datasets

Medical & Healthcare Verified

BloodMNIST

The BloodMNIST is based on a dataset of individual normal cells, captured from individuals without infection, hematologic or oncologic disease and free of any pharmacologic treatment…

Samples
17,092
Type
Images
Licence
CC-BY-4.0
Source
Zenodo
Medical & Healthcare Restricted access

NIH ChestX-ray14

The NIH ChestX-ray14 dataset contains 112,120 frontal-view chest X-ray images from 30,805 unique patients. Each image may carry one or more of 14 disease labels that…

Samples
112,120
Type
Images
Licence
Unknown
Source
NIH Clinical Center
Medical & Healthcare Verified

ChestMNIST

The ChestMNIST is based on the NIH-ChestXray14 dataset, a dataset comprising 112,120 frontal-view X-Ray images of 30,805 unique patients with the text-mined 14 disease labels, which…

Samples
112,120
Type
Images
Licence
CC-BY-4.0
Source
Zenodo
Medical & Healthcare Verified

AdrenalMNIST3D

The AdrenalMNIST3D is a new 3D shape classification dataset, consisting of shape masks from 1,584 left and right adrenal glands (i.e., 792 patients). Collected from Zhongshan…

Samples
1,584
Type
Images
Licence
CC-BY-4.0
Source
Zenodo
Medical & Healthcare Verified

OrganSMNIST

The OrganSMNIST is based on 3D computed tomography (CT) images from Liver Tumor Segmentation Benchmark (LiTS). It is renamed from OrganMNIST_Sagittal (in MedMNIST v1) for simplicity.

Samples
25,211
Type
Images
Licence
CC-BY-4.0
Source
Zenodo
Medical & Healthcare Verified

BreastMNIST

The BreastMNIST is based on a dataset of 780 breast ultrasound images. It is categorized into 3 classes: normal, benign, and malignant.

Samples
780
Type
Images
Licence
CC-BY-4.0
Source
Zenodo