An application for detecting skin diseases using advanced Vision Transformer models. The model is quantized with ONNX for efficient performance and is served via a FastAPI backend.
The dataset used for this project is publicly available on Kaggle:
Skin Disease Dataset
This dataset consists of images classified into 22 distinct skin disease categories.
- Model Architecture: Vision Transformer (ViT)
- Model Optimization: Quantized using ONNX for faster inference and reduced resource usage.
The model can detect and classify the following skin conditions:
- Acne
- Actinic Keratosis
- Benign Tumors
- Bullous
- Candidiasis
- Drug Eruption
- Eczema
- Infestations/Bites
- Lichen
- Lupus
- Moles
- Psoriasis
- Rosacea
- Seborrheic Keratoses
- Skin Cancer
- Sun/Sunlight Damage
- Tinea
- Unknown/Normal
- Vascular Tumors
- Vasculitis
- Vitiligo
- Warts
The backend is built with FastAPI, enabling seamless API interactions.
Skin Disease Detect API Documentation
Ensure you have Python and pip installed. Then install the required packages:
pip install -r requirements.txtTo start the application locally, run the following command:
gunicorn service.main:app --workers 2 --worker-class uvicorn.workers.UvicornWorkerDeveloped and maintained by Prashant Kumar Mishra.
GitHub: pacificrm
LinkedIn: Profile Link
For queries, suggestions, or contributions, feel free to reach out!