Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets
Abstract The remarkable progress of deep learning in dermatological tasks has brought us closer to achieving diagnostic accuracies comparable to those of human experts. However, while large datasets play a crucial role in the development of reliable deep neural network models, the quality of data th...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-02-01
|
| Series: | Scientific Data |
| Online Access: | https://doi.org/10.1038/s41597-025-04382-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832572021079277568 |
|---|---|
| author | Kumar Abhishek Aditi Jain Ghassan Hamarneh |
| author_facet | Kumar Abhishek Aditi Jain Ghassan Hamarneh |
| author_sort | Kumar Abhishek |
| collection | DOAJ |
| description | Abstract The remarkable progress of deep learning in dermatological tasks has brought us closer to achieving diagnostic accuracies comparable to those of human experts. However, while large datasets play a crucial role in the development of reliable deep neural network models, the quality of data therein and their correct usage are of paramount importance. Several factors can impact data quality, such as the presence of duplicates, data leakage across train-test partitions, mislabeled images, and the absence of a well-defined test partition. In this paper, we conduct meticulous analyses of three popular dermatological image datasets: DermaMNIST, its source HAM10000, and Fitzpatrick17k, uncovering these data quality issues, measure the effects of these problems on the benchmark results, and propose corrections to the datasets. Besides ensuring the reproducibility of our analysis, by making our analysis pipeline and the accompanying code publicly available, we aim to encourage similar explorations and to facilitate the identification and addressing of potential data quality issues in other large datasets. |
| format | Article |
| id | doaj-art-9fcad7aa76cf4ee5be5b308c695e9004 |
| institution | Kabale University |
| issn | 2052-4463 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Data |
| spelling | doaj-art-9fcad7aa76cf4ee5be5b308c695e90042025-02-02T12:08:10ZengNature PortfolioScientific Data2052-44632025-02-0112112110.1038/s41597-025-04382-5Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image DatasetsKumar Abhishek0Aditi Jain1Ghassan Hamarneh2School of Computing Science, Simon Fraser UniversityDepartment of Mathematics, Indian Institute of Technology DelhiSchool of Computing Science, Simon Fraser UniversityAbstract The remarkable progress of deep learning in dermatological tasks has brought us closer to achieving diagnostic accuracies comparable to those of human experts. However, while large datasets play a crucial role in the development of reliable deep neural network models, the quality of data therein and their correct usage are of paramount importance. Several factors can impact data quality, such as the presence of duplicates, data leakage across train-test partitions, mislabeled images, and the absence of a well-defined test partition. In this paper, we conduct meticulous analyses of three popular dermatological image datasets: DermaMNIST, its source HAM10000, and Fitzpatrick17k, uncovering these data quality issues, measure the effects of these problems on the benchmark results, and propose corrections to the datasets. Besides ensuring the reproducibility of our analysis, by making our analysis pipeline and the accompanying code publicly available, we aim to encourage similar explorations and to facilitate the identification and addressing of potential data quality issues in other large datasets.https://doi.org/10.1038/s41597-025-04382-5 |
| spellingShingle | Kumar Abhishek Aditi Jain Ghassan Hamarneh Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets Scientific Data |
| title | Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets |
| title_full | Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets |
| title_fullStr | Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets |
| title_full_unstemmed | Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets |
| title_short | Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets |
| title_sort | investigating the quality of dermamnist and fitzpatrick17k dermatological image datasets |
| url | https://doi.org/10.1038/s41597-025-04382-5 |
| work_keys_str_mv | AT kumarabhishek investigatingthequalityofdermamnistandfitzpatrick17kdermatologicalimagedatasets AT aditijain investigatingthequalityofdermamnistandfitzpatrick17kdermatologicalimagedatasets AT ghassanhamarneh investigatingthequalityofdermamnistandfitzpatrick17kdermatologicalimagedatasets |