Descriptor : The HAM 10000 dataset , a large collection of multi-source dermatoscopic images of common pigmented skin lesions
By P. Tschandl, C. Rosendahl, Harald Kittler. 2018
3,654 citations recorded by the source corpus.
Training of neural networks for automated diagnosis of pigmented skin lesions is hampered by the small size and lack of diversity of available datasets of dermatoscopic images. We tackle this problem by releasing the HAM10000 (“Human Against Machine with 10000 training images”) dataset. We collected dermatoscopic images from different populations acquired and stored by different modalities. Given this diversity we had to apply different acquisition and cleaning methods and developed semi-automatic workflows utilizing specifically trained neural networks. The final dataset consists of 10015 dermatoscopic images which are released as a training set for academic machine learning purposes and are publicly available through the ISIC archive. This benchmark dataset can be used for machine learning and for comparisons with human experts. Cases include a representative collection of all important diagnostic categories in the realm of pigmented lesions. More than 50% of lesions have been confirmed by pathology, while the ground truth for the rest of the cases was either follow-up, expert consensus, or confirmation by in-vivo confocal microscopy. Design Type(s) database creation objective • data integration objective • image format conversion objective Measurement Type(s) skin lesions Technology Type(s) digital curation Factor Type(s) diagnosis • Diagnostic Procedure • age • biological sex • animal body part Sample Characteristic(s) Homo sapiens • skin of body Design Type(s) database creation objective • data integration objective • image format conversion objective Measurement Type(s) skin lesions Technology Type(s) digital curation Factor Type(s) diagnosis • Diagnostic Procedure • age • biological sex • animal body part Sample Characteristic(s) Homo sapiens • skin of body Machine-accessible metadata file describing the reported data (ISA-Tab format)
Subfields: Artificial Intelligence, Epidemiology, Oncology, Medical Imaging
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Lucy R. Chai
- Fahad Shahbaz Khan
- Salman Hameed Khan
- A. Ben Hamza
- Hasib Zunair
- Bingbing Ni
- Neel Guha
- Rogério Schmidt Feris
In Epidemiology
- Lucy R. Chai
- Mohammad Sohel Rahman
- A. Ben Hamza
- Nabil Ibtehaz
- Hasib Zunair
- Nabila Abraham
- Naimul Mefraz Khan
- Virginia Smith
In Oncology
- Lucy R. Chai
- Noel C. F. Codella
- Huan Fu
- Jiancheng Yang
- Bingbing Ni
- Debesh Jha
- Pål Halvorsen
- Michael Alexander Riegler
In Medical Imaging
- Lucy R. Chai
- Jiancheng Yang
- Bingbing Ni
- Mohammad Sohel Rahman
- Virginia Smith
- Neel Guha
- A. Ben Hamza
- Nabil Ibtehaz
Citations in the corpus
- Florence: A New Foundation Model for Computer Vision (2021, neutral): “…leaf images, 38 disease states over 14 plant species), EuroSAT (Helber et al., 2019) (RGB satellite images, 10 categories), ISIC 2018 (Codella et al., 2019; Tschandl et al., 2018) (dermoscopic images of skin lesions, 7 disease states), and ChestX (Wang et al., 2017) (Chest X-rays, 16 conditions).”
- QUOTIENT: Two-Party Secure Neural Network Training and Prediction (2019, neutral): “Five of these are privacy sensitive in nature, as they include health datasets—Thyroid [45], Breast cancer [8], MotionSense [38], Skin cancer MNIST [53] and a financial credit dataset—German credit [12].”
- Skin lesion classification using ensembles of multi-resolution EfficientNets with meta data (2019, positive): “Furthermore, the training dataset, previously HAM10000 [16], was extended by additional data from the BCN 20000 [5] and MSK dataset [4].”
- DCSAU-Net: A Deeper and More Compact Split-Attention U-Net for Medical Image Segmentation (2022, methodology): “3) Experimental analysis is conducted with four different medical imagesegmentationdatasets, including2018Data Science Bowl [14], ISIC-2018Lesion BoundarySegmen-tation [ 15, 16], CVC-ClinicDB [17], and a multi-class segmentation dataset: SegPC-2021 [18].”
- Characteristics of publicly available skin cancer image datasets: a systematic review. (2021, background): “42,44,47 However, selection bias can occur where specific groups are excluded (eg, facial images or identifiable tattoos).”
- U-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications (2020, background): “Examples of Dermoscopy images from the ISIC 2018 challenge [274], [275].”
- A Novel Focal Tversky Loss Function With Improved Attention U-Net for Lesion Segmentation (2018, methodology): “Experiments were performed on the Breast Ultrasound Lesions 2017 dataset B (BUS) [8] and the ISIC 2018 skin lesion dataset [9], [10], both datasets suffering from class imbalance and large intra class variation in lesion sizes.”
- Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC) (2019, background): “For Part 3, 10,015 dermoscopic images with 7 ground truth classification labels were provided for training [10].”
- PAD-UFES-20: A skin lesion dataset composed of patient data and clinical images collected from smartphones (2020, result): “This number is compatible with other skin lesion datasets described in literature, for instance, the HAM1000 9 has 53.3% of biopsy-proven samples.”
- Eavesdrop the Composition Proportion of Training Labels in Federated Learning (2019, background): “and use the rest for testing. We choose Resnet18 as the learning model for Fer2013. It is trained under the setting L bs = 64, L lr = 0:02, L ep = 3. The testing accuracy is 95%. HAM10000. HAM10000 [41] is a large collection of multi-source dermatoscopic images of pigmented lesions. There are nearl”
- SplitFed: When Federated Learning Meets Split Learning (2020, background): “, the Human Against Machine with 10000 training images [19].”
- BCN20000: Dermoscopic Lesions in the Wild (2019, background): “However, as the authors already pointed out, the algorithms tended to perform worse on images from other dermoscopic data sources, which were not represented in the HAM10000 dataset [12].”
- MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification (2021, methodology): “The DermaMNIST is based on the HAM10000 19–21 , a large collection of multi-source dermatoscopic images of common pigmented skin lesions.”
- A New Benchmark for Evaluation of Cross-Domain Few-Shot Learning (2019, methodology): “The proposed benchmark includes data from the CropDiseases [23], EuroSAT [13], ISIC2018 [32, 5], and ChestX [35] datasets, which covers plant disease images, satellite images, dermoscopic images of skin lesions, and X-ray images, respectively.”