Skip to content
Open Imaging Index
Photography Lower limbSkinTrunkUpper limb

ISIC 2020

SIIM-ISIC 2020 Melanoma Classification Challenge Dataset

44,108 dermoscopic images of skin lesions from 2,746 patients at six centres (33,126 labelled training images of 2,056 patients, 584 of them histopathologically confirmed melanomas), with approximate age, sex and anatomic site.

Go to dataset Edit Report

Overview

The ISIC 2020 dataset was built for the SIIM-ISIC Melanoma Classification challenge, which ran on Kaggle from May to August 2020. Unlike earlier ISIC challenge sets, it groups lesions by patient, so a model can compare a suspicious lesion with the other lesions on the same person ("ugly duckling" reasoning). It is used for melanoma classification and for studying patient-level context.

Composition

The labelled training part holds 33,126 dermoscopic images of 2,056 patients; 584 images show melanoma and 428 patients have at least one. The remaining lesions are benign, either confirmed by biopsy or by at least six months of follow-up without change. A separate test part of 10,982 images from 690 patients is downloadable, but its labels were never released. Each image comes with an anonymised patient id, approximate age (rounded to five years), sex and a coarse anatomic site; version 2 of the training table adds a lesion id. Images are offered as JPEG and as DICOM with embedded metadata.

Acquisition

Images come from Memorial Sloan Kettering Cancer Center, Hospital Clínic de Barcelona, the Medical University of Vienna, Melanoma Institute Australia and the Sydney Melanoma Diagnosis Centre, and The University of Queensland, mostly high-risk or referral clinics. The test set also includes cases from a hospital in Athens. Contact and non-contact, polarised and non-polarised dermoscopy are mixed; the capture devices were not recorded.

Annotations

Malignant labels were checked against histopathology reports and visually by a dermoscopy expert. Melanoma in situ and invasive melanoma share one label. Most benign lesions carry the diagnosis "unknown"; nevus, seborrheic keratosis, lentigo and a few other benign diagnoses are given where available.

Known limitations

  • 425 training images are pixel-identical duplicates; a list is provided, and lesion ids in version 2 help remove them.
  • Darker skin types are under-represented, and melanoma is far more frequent than in the general population.
  • Each lesion is one image with one type of dermoscopy.
  • Age bins on this page are approximate because ages are rounded to five years.
  • Condition counts cover training labels only.

Cohort

Aggregate numbers from the sources below. Bars are relative to the 2,746 subjects.

Sex

  • Female 1,303 47%
  • Male 1,441 52%
  • Unknown 2 0%

Age

mean 51.3
0
200
400
600
~641
0-910-1920-2930-3940-4950-5960-6970-7980-8990+

Age by sex

Reported cross table. Missing cells were not published (fewer than 10 subjects or not reported).

Female Male
  • 21
    80-89
    51
  • 90
    70-79
    195
  • 204
    60-69
    254
  • 320
    50-59
    313
  • 343
    40-49
    298
  • 215
    30-39
    205
  • 86
    20-29
    97
  • 17
    10-19
    21

Condition

subjects, values can overlap

  • Melanocytic nevus 878 32%
  • Melanoma 428 16%
  • Seborrheic keratosis 104 4%

Anatomy

subjects

  • Trunk 2,614 95%
  • Lower limb 2,174 79%
  • Upper limb 1,906 69%

Split

subjects

  • Training 2,056 75%
  • Test 690 25%

License and access

Our reading of the license, not legal advice. Before you use the data, read the original license and confirm that your use is allowed. We take no responsibility for how you use a dataset. Full disclaimer

Access
Open download

Download without an account

Access page

Creative Commons Attribution-NonCommercial 4.0 International

Use, share and adapt the data with credit, but only for non-commercial purposes.

The challenge data page lists the 2020 data under CC-BY-NC and asks that the aggregate data be cited as "International Skin Imaging Collaboration. SIIM-ISIC 2020 Challenge Dataset", DOI 10.34970/2020-ds01.

Original license text Version read: 4.0 Checked 2026-10-07

What you can do

  • No
  • Conditional
  • Yes
  • Yes

What you can share

  • Conditional
  • Conditional
  • Conditional

What you must do

  • Yes
  • Share alike No
  • No
  • No
  • Manuscript review No
  • Release code No
  • Return results No
  • Delete after use No

Limits

  • No
  • Location limits No

Citation

International Skin Imaging Collaboration. SIIM-ISIC 2020 Challenge Dataset. https://doi.org/10.34970/2020-ds01 (2020). Rotemberg V, Kurtansky N, et al. A patient-centric dataset of images and metadata for identifying melanomas using clinical context. Sci Data 8, 34 (2021).

All numbers

Every number on this page, as stored in stats.csv, with its source.

MeasureBreakdownValueSource
Subjectstotal
Unique patient IDs; training and test patients are disjoint
2,746isic2020-metadata-csv
Imagestotal
33126 training + 10982 test
44,108isic2020-metadata-csv
Subjectssplit=test 690isic2020-metadata-csv
Subjectssplit=train 2,056rotemberg2021
Table 1
Subjectssex=female 1,303isic2020-metadata-csv
Subjectssex=male 1,441isic2020-metadata-csv
Subjectssex=unknown 2isic2020-metadata-csv
Subjectsage=0-9
Age bins use the youngest age_approx per patient; age_approx is rounded to 5 years
~1isic2020-metadata-csv
Subjectsage=10-19 ~38isic2020-metadata-csv
Subjectsage=20-29 ~183isic2020-metadata-csv
Subjectsage=30-39 ~420isic2020-metadata-csv
Subjectsage=40-49 ~641isic2020-metadata-csv
Subjectsage=50-59 ~633isic2020-metadata-csv
Subjectsage=60-69 ~458isic2020-metadata-csv
Subjectsage=70-79 ~285isic2020-metadata-csv
Subjectsage=80-89 ~72isic2020-metadata-csv
Subjectsage=90+ ~12isic2020-metadata-csv
Subjectssex=female;age=10-19 ~17isic2020-metadata-csv
Subjectssex=female;age=20-29 ~86isic2020-metadata-csv
Subjectssex=female;age=30-39 ~215isic2020-metadata-csv
Subjectssex=female;age=40-49 ~343isic2020-metadata-csv
Subjectssex=female;age=50-59 ~320isic2020-metadata-csv
Subjectssex=female;age=60-69 ~204isic2020-metadata-csv
Subjectssex=female;age=70-79 ~90isic2020-metadata-csv
Subjectssex=female;age=80-89 ~21isic2020-metadata-csv
Subjectssex=male;age=10-19 ~21isic2020-metadata-csv
Subjectssex=male;age=20-29 ~97isic2020-metadata-csv
Subjectssex=male;age=30-39 ~205isic2020-metadata-csv
Subjectssex=male;age=40-49 ~298isic2020-metadata-csv
Subjectssex=male;age=50-59 ~313isic2020-metadata-csv
Subjectssex=male;age=60-69 ~254isic2020-metadata-csv
Subjectssex=male;age=70-79 ~195isic2020-metadata-csv
Subjectssex=male;age=80-89 ~51isic2020-metadata-csv
Subjectssex=female;split=test 326isic2020-metadata-csv
Subjectssex=female;split=train 977rotemberg2021
Table 1
Subjectssex=male;split=test 364isic2020-metadata-csv
Subjectssex=male;split=train 1,077rotemberg2021
Table 1
Imagessplit=test 10,982isic2020-metadata-csv
Imagessplit=train 33,126rotemberg2021
Table 1 (lesions)
Imagessex=female 20,708isic2020-metadata-csv
Imagessex=male 23,335isic2020-metadata-csv
Imagessex=unknown 65isic2020-metadata-csv
Imagesage=0-9
Age bins use age_approx of the image (rounded to 5 years)
~2isic2020-metadata-csv
Imagesage=10-19 ~197isic2020-metadata-csv
Imagesage=20-29 ~2,684isic2020-metadata-csv
Imagesage=30-39 ~6,942isic2020-metadata-csv
Imagesage=40-49 ~11,081isic2020-metadata-csv
Imagesage=50-59 ~10,589isic2020-metadata-csv
Imagesage=60-69 ~7,555isic2020-metadata-csv
Imagesage=70-79 ~3,965isic2020-metadata-csv
Imagesage=80-89 ~931isic2020-metadata-csv
Imagesage=90+ ~94isic2020-metadata-csv
Imagessex=female;age=10-19 ~108isic2020-metadata-csv
Imagessex=female;age=20-29 ~1,490isic2020-metadata-csv
Imagessex=female;age=30-39 ~3,824isic2020-metadata-csv
Imagessex=female;age=40-49 ~5,052isic2020-metadata-csv
Imagessex=female;age=50-59 ~5,308isic2020-metadata-csv
Imagessex=female;age=60-69 ~3,462isic2020-metadata-csv
Imagessex=female;age=70-79 ~1,095isic2020-metadata-csv
Imagessex=female;age=80-89 ~306isic2020-metadata-csv
Imagessex=female;age=90+ ~60isic2020-metadata-csv
Imagessex=male;age=10-19 ~89isic2020-metadata-csv
Imagessex=male;age=20-29 ~1,194isic2020-metadata-csv
Imagessex=male;age=30-39 ~3,118isic2020-metadata-csv
Imagessex=male;age=40-49 ~6,029isic2020-metadata-csv
Imagessex=male;age=50-59 ~5,281isic2020-metadata-csv
Imagessex=male;age=60-69 ~4,093isic2020-metadata-csv
Imagessex=male;age=70-79 ~2,870isic2020-metadata-csv
Imagessex=male;age=80-89 ~625isic2020-metadata-csv
Imagessex=male;age=90+ ~34isic2020-metadata-csv
Imagessex=female;split=test 4,727isic2020-metadata-csv
Imagessex=female;split=train 15,981isic2020-metadata-csv
Imagessex=male;split=test 6,255isic2020-metadata-csv
Imagessex=male;split=train 17,080isic2020-metadata-csv
Imagessex=unknown;split=train 65isic2020-metadata-csv
Subjectsanatomy=trunk
Patients with at least one image at this site (torso)
2,614isic2020-metadata-csv
Subjectsanatomy=upper_limb 1,906isic2020-metadata-csv
Subjectsanatomy=lower_limb 2,174isic2020-metadata-csv
Imagesanatomy=trunk
anatom_site_general = torso
22,692isic2020-metadata-csv
Imagesanatomy=upper_limb 6,556isic2020-metadata-csv
Imagesanatomy=lower_limb 10,918isic2020-metadata-csv
Imagescondition=melanoma
Training labels only; test labels are not public
584rotemberg2021
Table 1
Subjectscondition=melanoma
Training patients with at least one melanoma; test labels are not public
428rotemberg2021
Table 1
Imagescondition=melanocytic_nevus
Training labels only; most benign lesions are labelled unknown
5,193isic2020-metadata-csv
Subjectscondition=melanocytic_nevus
Training labels only
878isic2020-metadata-csv
Imagescondition=seborrheic_keratosis 135isic2020-metadata-csv
Subjectscondition=seborrheic_keratosis 104isic2020-metadata-csv
Subjectssex=unknown;split=train 2rotemberg2021
Table 1
Mean agetotal
Training patients only
51.3rotemberg2021
Table 1

Sources

The keys used in the table above.