You can deploy this model and start earning money today!

zrn2003/
skinsight-ai

Name: skinsight-ai
Availability: InStock

Views

No views yet

SkinSight AI - Skin Disease Detection

AI-powered skin disease analysis using deep learning. Upload a dermoscopic image to get instant diagnosis of skin conditions including Melanoma and Tinea.

🚀 Features

Two-Stage Pipeline: First filters non-skin images, then diagnoses valid skin conditions
Advanced Preprocessing: Digital hair removal, CLAHE enhancement, color normalization
Ensemble Deep Learning: Combines ResNet50 and InceptionV3 for high accuracy
Real-time Inference: Fast API responses for immediate results

📡 API Endpoints

Health Check

GET /ping

Response:

json

{
  "status": "alive",
  "filter_loaded": true,
  "fusion_loaded": true,
  "device": "cuda"
}

Prediction

bash

POST /predict
Content-Type: multipart/form-data

Request: Upload an image file

Response:

json

{
  "class": "Melanoma",
  "confidence": 0.95,
  "stage": "Diagnosis",
  "filter_check": "Passed (Skin)",
  "filter_confidence": 0.98
}

💻 Example Usage

Python

python

import requests

# Health check
response = requests.get("https://zrn2003-skinsight-ai.hf.space/ping")
print(response.json())

# Prediction
with open("skin_image.jpg", "rb") as f:
    files = {"file": f}
    response = requests.post(
        "https://zrn2003-skinsight-ai.hf.space/predict",
        files=files
    )
    result = response.json()
    print(f"Diagnosis: {result['class']}")
    print(f"Confidence: {result['confidence']:.2%}")

cURL

bash

# Health check
curl https://zrn2003-skinsight-ai.hf.space/ping

# Prediction
curl -X POST https://zrn2003-skinsight-ai.hf.space/predict \
  -F "file=@skin_image.jpg"

JavaScript/TypeScript

javascript

const formData = new FormData();
formData.append('file', fileInput.files[0]);

const response = await fetch('https://zrn2003-skinsight-ai.hf.space/predict', {
  method: 'POST',
  body: formData
});

const result = await response.json();
console.log(result);

🔬 Model Architecture

Filter Model: ResNet18 - Filters out non-skin images
Fusion Ensemble: ResNet50 + InceptionV3 - Diagnoses skin conditions
Classes: Melanoma, Tinea

📊 Performance

Filter Accuracy: ~100% on test samples
Diagnostic Accuracy: High confidence (>90%) on correct predictions
Latency: Real-time inference on GPU

⚠️ Medical Disclaimer

This tool is for research and educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of qualified healthcare providers with any questions regarding medical conditions.

📚 Citation

If you use this model in your research, please cite:

bibtex

@software{skinsight2025,
  title={SkinSight AI: Deep Learning for Skin Disease Detection},
  author={Your Name},
  year={2025},
  url={https://huggingface.co/zrn2003/SkinSight}
}

🔗 Links

Model Repository: zrn2003/SkinSight
Space API: https://zrn2003-skinsight-ai.hf.space

zrn2003/skinsight-ai