Skin Diseases Detection using Deep Learning

Overview

This project implements a deep learning solution for detecting and classifying various skin diseases from images. Using advanced convolutional neural networks, the system can help in early detection and classification of skin conditions, potentially assisting healthcare professionals in diagnosis.

Dataset Details

The dataset contains 34 different classes of skin diseases with a total of approximately 15,000 images. Each class represents a different skin condition with both English and Hindi names.

Classes and Image Distribution

1. Eczema (सूजन) - 500 images
2. Melanoma (त्वचा कैंसर) - 1000 images
3. Atopic Dermatitis (एक्ज़िमा खुजलीदार त्वचा की बीमारी) - 500 images
4. Basal Cell Carcinoma (कम घातक कैंसर) - 1000 images
5. Melanocytic Nevi (तिल) - 500 images
6. Benign Keratosis-like Lesions (गैर-हानिकारक घाव) - 500 images
7. Psoriasis pictures Lichen Planus (पपड़ीदार धब्बे) - 500 images
8. Seborrheic Keratoses and other Benign Tumors (त्वचा ट्यूमर) - 500 images
9. Tinea Ringworm Candidiasis (दाद) - 500 images
10. Warts Molluscum and Viral Infections (मस्से) - 500 images
11. Pigment (रंगद्रव्य) - 500 images
12. Enfeksiyonel (संक्रमणजन्य रोग) - 500 images
13. Akne (मुहाँसे) - 322 images
14. Enfeksiyonel (छूत) - 500 images
15. Actinic Keratosis and Malignant Lesions (पूर्व-कैंसर त्वचा) - 500 images
16. Bullous Disease (फफोले वाले त्वचा रोग) - 448 images
17. Cellulitis Impetigo and Bacterial Infections - 288 images
18. Exanthems and Drug Eruptions (त्वचा पर दाने) - 404 images
19. Hair Loss, Alopecia and Hair Diseases - 239 images
20. Herpes, HPV and STDs - 405 images
21. Light Diseases and Pigmentation Disorders - 568 images
22. Lupus and Connective Tissue diseases (संयोजी ऊतक की बीमारियाँ) - 420 images
23. Nail Fungus and Nail Diseases - 500 images
24. Poison Ivy and Contact Dermatitis (त्वचा की जलन) - 260 images
25. Rosacea (गुलाबी फुंसी) - 500 images
26. Scabies, Lyme Disease and Bites (किट-पतंग के काटने) - 431 images
27. Seborrheic Keratoses and Benign Tumors (सौम्य त्वचा के ट्यूमर) - 500 images
28. Systemic Disease (संपूर्ण शरीर) - 606 images
29. Fungal Infections (फंगल संक्रमण) - 500 images
30. Urticaria Hives (छींक-धब्बे) - 212 images
31. Vascular Tumors (रक्त वाहिकाओं के ट्यूमर) - 482 images
32. Vasculitis (रक्त वाहिका सूजन) - 416 images
33. Viral Infections (वायरल संक्रमण) - 500 images
34. Normal Skin - 1000 images

Features

• Automated skin disease detection from images
• Multi-class classification of various skin conditions
• Deep learning-based image analysis
• Interactive Jupyter notebook implementation
• High accuracy in disease classification

Technologies Used

• Python
• TensorFlow/Keras
• Jupyter Notebook
• NumPy
• Pandas
• Matplotlib/Seaborn for visualization

Installation

1. Clone the repository:

git clone https://github.com/itsluckysharma01/Skin_Diseases_Detection_Using_Deep-Learning_.git

2. Install required dependencies:

pip install -r requirements.txt

Usage

1. Open the Jupyter notebook:
   • Navigate to the Files directory
   • Open Skin_Diseases_Detection_Using_Deep_Learning(NOTEBOOK).ipynb
2. Follow the step-by-step instructions in the notebook to:
   • Load and preprocess the dataset
   • Train the model
   • Evaluate results
   • Make predictions

Project Structure

Files/
├── requirements.txt
└── Skin_Diseases_Detection_Using_Deep_Learning(NOTEBOOK).ipynb

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Author

• Lucky Sharma (@itsluckysharma01)

Acknowledgments

• Thanks to all contributors who helped in developing this project
• Special thanks to the medical professionals who helped in validating the results

Contact

• Lucky Sharma (@itsluckysharma01)

---

Gmail:- itsluckysharma001@gmail.com