Spaces:
Sleeping
Sleeping
Upload README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
title: PulmoScanAI
|
| 3 |
+
emoji: π«
|
| 4 |
+
colorFrom: blue
|
| 5 |
+
colorTo: green
|
| 6 |
+
sdk: docker
|
| 7 |
+
app_file: app.py
|
| 8 |
+
pinned: false
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# PulmoScanAI - AI Lung Cancer Detection System
|
| 12 |
+
|
| 13 |
+
An advanced web-based application for detecting lung cancer from histopathology images using a deep learning CNN model with feature-based analysis.
|
| 14 |
+
|
| 15 |
+
## Features
|
| 16 |
+
|
| 17 |
+
- **Real-time AI Analysis**: Uses TensorFlow/Keras deep learning model
|
| 18 |
+
- **Feature-based Detection**: Analyzes darkness, purple staining, and edge density
|
| 19 |
+
- **Beautiful UI**: Modern, responsive design with animated backgrounds
|
| 20 |
+
- **Drag & Drop Upload**: Easy image upload with preview
|
| 21 |
+
- **Confidence Score**: Displays detection confidence percentage
|
| 22 |
+
- **CORS Enabled**: Seamless frontend-backend communication
|
| 23 |
+
|
| 24 |
+
## How It Works
|
| 25 |
+
|
| 26 |
+
1. **Upload Image**: Drag & drop a histopathology image
|
| 27 |
+
2. **CNN Processing**: Model analyzes tissue patterns
|
| 28 |
+
3. **Feature Analysis**: Evaluates darkness, staining, and texture
|
| 29 |
+
4. **Result**: Shows diagnosis with confidence score
|
| 30 |
+
- **Green**: Normal tissue detected
|
| 31 |
+
- **Red**: Cancer detected
|
| 32 |
+
|
| 33 |
+
## API Endpoints
|
| 34 |
+
|
| 35 |
+
### Health Check
|
| 36 |
+
```
|
| 37 |
+
GET /api/health
|
| 38 |
+
```
|
| 39 |
+
|
| 40 |
+
### Prediction
|
| 41 |
+
```
|
| 42 |
+
POST /api/predict
|
| 43 |
+
Content-Type: multipart/form-data
|
| 44 |
+
```
|
| 45 |
+
|
| 46 |
+
**Request**: Image file in multipart form data
|
| 47 |
+
**Response**:
|
| 48 |
+
```json
|
| 49 |
+
{
|
| 50 |
+
"is_cancer": false,
|
| 51 |
+
"confidence": 0.92,
|
| 52 |
+
"diagnosis": "No Cancer Found",
|
| 53 |
+
"confidence_percentage": 92.0
|
| 54 |
+
}
|
| 55 |
+
```
|
| 56 |
+
|
| 57 |
+
## Model Information
|
| 58 |
+
|
| 59 |
+
- **Architecture**: Convolutional Neural Network (CNN)
|
| 60 |
+
- **Input**: 150Γ150 RGB images
|
| 61 |
+
- **Output**: 3-class classification (Adenocarcinoma, Normal, Squamous Cell Carcinoma)
|
| 62 |
+
- **Framework**: TensorFlow 2.13.0 / Keras
|
| 63 |
+
|
| 64 |
+
## Technical Stack
|
| 65 |
+
|
| 66 |
+
- **Frontend**: HTML5, CSS3, JavaScript (Vanilla)
|
| 67 |
+
- **Backend**: Python Flask with Flask-CORS
|
| 68 |
+
- **ML Framework**: TensorFlow 2.x / Keras
|
| 69 |
+
- **Image Processing**: OpenCV, Pillow, NumPy
|
| 70 |
+
|
| 71 |
+
## Project Structure
|
| 72 |
+
|
| 73 |
+
```
|
| 74 |
+
βββ app.py # Flask backend server
|
| 75 |
+
βββ best_lung_model.h5 # Trained CNN model
|
| 76 |
+
βββ PulmoScanAI.html # Web frontend
|
| 77 |
+
βββ requirements.txt # Python dependencies
|
| 78 |
+
βββ Dockerfile # Container configuration
|
| 79 |
+
βββ README.md # This file
|
| 80 |
+
```
|
| 81 |
+
|
| 82 |
+
## License
|
| 83 |
+
|
| 84 |
+
Β© 2025 PulmoScanAI β’ Next-Gen AI Pathology Platform
|