Skin Disease Classifier is an AI-powered web application that predicts skin conditions from images and returns rich metadata including symptoms, causes, treatments, language-translated health tips, and text-to-speech audio guidance.

• ✅ Classifies 23 common skin diseases using image input
• ✅ Provides English, Hausa, Yoruba, and Igbo health tips
• ✅ Speaks the diagnosis and advice using text-to-speech (TTS)
• ✅ Uses metadata for detailed disease descriptions, symptoms, causes, and treatments
• ✅ Built with TensorFlow (MobileNetV2), Gradio, and deployed on Hugging Face Spaces

👉 Launch on Hugging Face

• Architecture: EfficientNetB0 via Transfer Learning
• Dataset: DermNet skin disease images (23 classes)
• Framework: TensorFlow/Keras
• Deployment: Hugging Face Spaces with Gradio UI

# 1. Clone this repo
git clone https://github.com/your-username/skin-disease-classifier.git
cd skin-disease-classifier

# 2. Install required packages
pip install -r requirements.txt

# 3. Run the Gradio app
python app.py  # or gradio_app.py

This project would not have been possible without the contributions and tools from the following communities and platforms:

• DermNet NZ – For providing the dermatology image dataset that powers this classifier.
• TensorFlow – For deep learning and model development tools.
• Hugging Face – For model and app hosting through Spaces and huggingface_hub.
• Gradio – For building the interactive and user-friendly web interface.
• gTTS (Google Text-to-Speech) – For enabling voice output in multiple languages.
• Deep Translator – For enabling text translation into Hausa, Yoruba, and Igbo.
• Pillow (PIL) – For handling image inputs.
• NumPy – For efficient array operations used in preprocessing and predictions.

Special thanks to the open-source community for providing tools and inspiration, and to all healthcare professionals whose real-world insights guide the meaningful application of AI in medicine.