Table 1 List of existing image datasets related to the skin diseases in question
Name | Size in MB | Description | Ref | Task | Last Accessed |
|---|---|---|---|---|---|
Skin dataset | 664 | 8,212 on-the-wild images, including 1,566 VI images in JPEG format Categories are divided into seven classes | [114] | Classification | 1/4/2024 |
DermNet | 2,000 | Images of 23 types of skin diseases, in JPEG format. | [115] | Classification | 1/4/2024 |
VitMon | 2 | 36 grayscale images in JPEG format along with the corresponding lesion masks. | [116] | Segmentation | 1/4/2024 |
Vitiligo | 130 | 3,628 images in PNG format divided into two classes: healthy and VI affected. | [117] | Classification | 1/4/2024 |
vitiligo-detection 01 | 56.6 | 189 VI dermoscopy images in JPEG format, where the VI-affected area is annotated in a bounding box. | [118] | Object detection | 1/4/2024 |
vitiligo computer vision | 76.3 | 2,118 VI-on-the-wild images in JPEG format, including the relevant annotations as bounding boxes. | [119] | Object detection | 1/4/2024 |
hair image | 26.2 | 535 images with the corresponding segmentation masks. | [120] | Segmentation | 20/4/2024 |
VtigoDataset2 | 20.6 | 1,187 VI on-the-wild images in JPEG format, annotated with corresponding bounding box. Augmentation is already applied. | [121] | Object detection | 1/4/2024 |
vitiligo-seg-01 | 115 | 384 VI dermoscopy images in JPEG format with the corresponding segmentation masks. Augmentation is already applied. | [122] | Segmentation | 1/4/2024 |
Extent of Hair Loss in Patients with AA | 78 | 2,716 pixel-wise annotations used to train the hair loss identifier (mask), and the hair loss identifier (target). | [123] | Segmentation | 23/4/2024 |
Image Dataset | 17 | 290 images of top head images, and annotation of the head perimeter. | [124] | Segmentation | 25/4/2024 |
Hair image dataset | 26.8 | 534 images of top head images and annotation of the alopecia patches. | [125] | Segmentation | 25/4/2024 |
Eczema Disease Classification | 9.2 | 510 images. 5 classes, various body sites. | [126] | Classification | 25/4/2024 |
eczema Computer Vision Project | 78.1 | 1,512 images. Annotations using small boxes of eczema in various body sites. | [127] | Object detection | 25/4/2024 |
disease area detection Image Dataset | 274 | 1,440 Images. Classes: Acne, Eczema, Psoriasis, VI. | [128] | Instance Segmentation | 1/4/2024 |
New_UAE Computer Vision Project | 230 | 5,122 Images. Classes: Acne, Moll, Psoriasis. Annotation with a box of the affected area. | [129] | Object detection | 1/4/2024 |
AtopicDermatitis | 113 | 2,630 images from a close distance and various body sites labelled as AD or no AD. | [130] | Classification | 19/4/2024 |
Skin disease dataset | 52.2 | 1,147 images, 10 classes including AD, Eczema, and Psoriasis. | [131] | Object detection | 19/4/2024 |
3. Atopic Dermatitis | 2.62 | 52 AD images annotated at pixel-level. | [132] | Instance Segmentation | 19/4/2024 |
Nummular preprocessing dataset | 7.53 | 622 AD and Nummular Dermatitis images. | [133] | Classification | 19/4/2024 |
Skin_Disease_AK | 127 | 13,159 images over 20 classes, including 228 AD. Extensive duplication is observed. | [134] | Classification | 19/4/2024 |
FYP Eczema | 0.767 | 52 images with pixel-level annotations. | [135] | Instance Segmentation | 19/4/2024 |
Skin Disease Classification | 169 | 81 AD images, part of a 9-class dataset. | [136] | Classification | 19/4/2024 |
20 Skin Diseases Dataset | 321 | 3,056 images representing 20 classes, including AD and eczema. | [137] | Classification | 19/4/2024 |
Skin diseases image dataset | 6,000 | 10 classes including 1,257 AD images, 1,677 Eczema images, 2,055 Psoriasis images. | [138] | Classification | 19/4/2024 |