https://github.com/deveshsangwan/skin-lesions-classification
This repo includes classifier trained to distinct 7 type of skin lesions
https://github.com/deveshsangwan/skin-lesions-classification
convolutional-neural-networks deep-learning ensemble-learning ensemble-machine-learning ensemble-model ham10000 harvard-dataverse inception inception-v3 machine-learning resnet resnet-152 skin-cancer skin-lesion-classification skin-lesions vgg vgg19 xception xception-model
Last synced: 3 months ago
JSON representation
This repo includes classifier trained to distinct 7 type of skin lesions
- Host: GitHub
- URL: https://github.com/deveshsangwan/skin-lesions-classification
- Owner: deveshsangwan
- License: gpl-3.0
- Created: 2020-10-12T08:30:53.000Z (about 5 years ago)
- Default Branch: main
- Last Pushed: 2021-10-11T05:47:51.000Z (about 4 years ago)
- Last Synced: 2025-04-08T22:42:16.525Z (6 months ago)
- Topics: convolutional-neural-networks, deep-learning, ensemble-learning, ensemble-machine-learning, ensemble-model, ham10000, harvard-dataverse, inception, inception-v3, machine-learning, resnet, resnet-152, skin-cancer, skin-lesion-classification, skin-lesions, vgg, vgg19, xception, xception-model
- Language: Jupyter Notebook
- Homepage:
- Size: 962 KB
- Stars: 10
- Watchers: 1
- Forks: 7
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Skin-lesions-classification
This repo includes classifier trained to distinct 7 type of skin lesions.
## Dataset
Dataset consists of 10015 images which describe 7 type of lesions.
Dataset includes images of following 7 skin lesions:
* Actinic keratoses and intraepithelial carcinoma / Bowen's disease (akiec)
* basal cell carcinoma (bcc)
* benign keratosis-like lesions (solar lentigines / seborrheic keratoses and lichen-planus like keratoses, bkl)
* dermatofibroma (df)
* melanoma (mel)
* melanocytic nevi (nv)
* vascular lesions (angiomas, angiokeratomas, pyogenic granulomas and hemorrhage, vasc)
[Dataset link](https://www.kaggle.com/kmader/skin-cancer-mnist-ham10000)
## Model Training
The following CNN models are used:
* InceptionV3
* Resnet152
* VGG19
* Xception
* Ensemble
## Results
| Model | Accuracy |
| ----- | -------- |
| InceptionV3 | 0.862 |
| Resnet152 | 0.802 |
| VGG19 | 0.835 |
| Xception | 0.862 |
| Ensemble | 0.875 |
## Conclusion
Out of all the algorithms used, Ensemble of Xception, Resnet152, InceptionV3 gives the best accuracy of 0.8753.