DOWNLOAD_INSTRUCTIONS.md

# Download Instructions for Public Medical Imaging Datasets

<!--
Medical Imaging Dataset - Download Instructions
===============================================

Project: Medical Imaging Dataset
Website: https://rskworld.in
Contact: help@rskworld.in, support@rskworld.in
Phone: +91 93305 39277
Founder: Molla Samser
Designer & Tester: Rima Khatun
-->

## Important Notes

⚠️ **All datasets are for educational and research purposes only.**
- NOT for actual medical diagnosis or treatment
- Always check the license and terms of use
- Respect patient privacy and data usage guidelines
- Some datasets require registration or API keys

## Available Datasets

### 1. Chest X-Ray Images (Pneumonia)
- **Source**: Kaggle
- **URL**: https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia
- **Size**: ~1.2 GB
- **Images**: ~5,863 chest X-ray images
- **License**: CC BY 4.0
- **Download**: Requires Kaggle account and API key
- **Instructions**:
1. Create a Kaggle account at https://www.kaggle.com
2. Go to Account Settings → API → Create New API Token
3. Download the kaggle.json file
4. Install Kaggle API: `pip install kaggle`
5. Place kaggle.json in ~/.kaggle/ (or C:\Users\YourUsername\.kaggle\ on Windows)
6. Run: `kaggle datasets download -d paultimothymooney/chest-xray-pneumonia`
7. Extract and organize the images

### 2. NIH Chest X-ray Dataset
- **Source**: NIH (National Institutes of Health)
- **URL**: https://www.kaggle.com/datasets/nih-chest-xrays/data
- **Size**: ~40 GB
- **Images**: ~112,120 chest X-ray images
- **License**: Public Domain
- **Download**: Requires Kaggle account
- **Instructions**:
1. Use Kaggle API: `kaggle datasets download -d nih-chest-xrays/data`
2. Extract the large dataset
3. Organize by condition/diagnosis

### 3. Brain MRI Images for Brain Tumor Detection
- **Source**: Kaggle
- **URL**: https://www.kaggle.com/datasets/navoneel/brain-mri-images-for-brain-tumor-detection
- **Size**: ~300 MB
- **Images**: ~3,000+ brain MRI images
- **License**: CC BY 4.0
- **Download**: Requires Kaggle account
- **Instructions**:
1. Use Kaggle API: `kaggle datasets download -d navoneel/brain-mri-images-for-brain-tumor-detection`
2. Extract and organize

### 4. COVID-19 X-ray Dataset
- **Source**: GitHub
- **URL**: https://github.com/ieee8023/covid-chestxray-dataset
- **Size**: ~50 MB
- **Images**: ~100+ images
- **License**: MIT
- **Download**: Direct from GitHub
- **Instructions**:
1. Clone repository: `git clone https://github.com/ieee8023/covid-chestxray-dataset.git`
2. Or download ZIP from GitHub
3. Extract images from repository

### 5. Medical MNIST
- **Source**: Kaggle
- **URL**: https://www.kaggle.com/datasets/andrewmvd/medical-mnist
- **Size**: ~200 MB
- **Images**: ~66,000 images across multiple modalities
- **License**: CC BY 4.0
- **Download**: Requires Kaggle account
- **Instructions**:
1. Use Kaggle API: `kaggle datasets download -d andrewmvd/medical-mnist`
2. Extract and organize by modality

## Quick Setup Guide

### Install Kaggle API
```bash
pip install kaggle
```

### Setup Kaggle Credentials
1. Go to https://www.kaggle.com/account
2. Scroll to "API" section
3. Click "Create New API Token"
4. Download `kaggle.json`
5. Place it in:
- **Windows**: `C:\Users\YourUsername\.kaggle\kaggle.json`
- **Linux/Mac**: `~/.kaggle/kaggle.json`
6. Set permissions (Linux/Mac): `chmod 600 ~/.kaggle/kaggle.json`

### Download a Dataset
```bash
# Example: Download Chest X-Ray dataset
kaggle datasets download -d paultimothymooney/chest-xray-pneumonia

# Extract
unzip chest-xray-pneumonia.zip -d data/
```

## Organizing Downloaded Data

After downloading, organize the data into the project structure:

```
data/
├── xray/
│ ├── images/
│ └── labels/
├── ct_scan/
│ ├── images/
│ └── labels/
└── mri/
├── images/
└── labels/
```

## Attribution

When using these datasets, please:
- Cite the original dataset sources
- Include license information
- Follow the terms of use
- Give proper attribution to the creators

## Contact

- **Website**: [rskworld.in](https://rskworld.in)
- **Email**: help@rskworld.in, support@rskworld.in
- **Phone**: +91 93305 39277

## Additional Public Data Sources

### 6. The Cancer Imaging Archive (TCIA)
- **URL**: https://www.cancerimagingarchive.net/
- **Description**: Open-access database of medical images for cancer research
- **Modalities**: CT, MRI, PET scans
- **Access**: Free registration required
- **License**: Various (check individual collections)
- **Instructions**:
1. Register at the website
2. Browse collections by cancer type
3. Download via TCIA Data Portal or NBIA

### 7. Radiopaedia
- **URL**: https://radiopaedia.org/
- **Description**: Educational radiology resource with case studies
- **Modalities**: X-ray, CT, MRI, Ultrasound
- **Access**: Free, educational use
- **License**: CC BY-NC-SA
- **Instructions**: Browse cases and download images for educational purposes

### 8. OpenNeuro
- **URL**: https://openneuro.org/
- **Description**: Open-science database for brain imaging
- **Modalities**: MRI, fMRI, EEG
- **Access**: Free, open access
- **License**: Various (check individual datasets)
- **Instructions**:
1. Browse datasets
2. Download via web interface or API
3. Data in BIDS format

### 9. Stanford AIMI Shared Datasets
- **URL**: https://aimi.stanford.edu/resources/shared-datasets
- **Description**: Stanford AI in Medicine & Imaging shared datasets
- **Modalities**: Multiple imaging modalities
- **Access**: Free for research purposes
- **License**: Check individual datasets
- **Instructions**: Browse and download from the website

### 10. Medical Imaging Data Resource Center (MIDRC)
- **URL**: https://www.midrc.org/
- **Description**: Medical Imaging Data Resource Center (COVID-19 focus)
- **Modalities**: CT, X-ray, MRI
- **Access**: Free registration required
- **License**: Various
- **Instructions**: Register and access via the portal

### 11. MedMNIST v2
- **URL**: https://medmnist.com/
- **Description**: Large-scale standardized biomedical images
- **Modalities**: 12 2D + 6 3D datasets covering various modalities
- **Access**: Free, direct download
- **License**: CC BY 4.0
- **Instructions**:
1. Visit the website
2. Download datasets directly
3. Pre-processed to 28x28 (2D) or 28x28x28 (3D)

### 12. ReXGradient-160K
- **URL**: Check arXiv paper: https://arxiv.org/abs/2505.00228
- **Description**: 160,000 chest X-ray studies with radiological reports
- **Modalities**: Chest X-rays
- **Access**: Research use (check availability)
- **License**: Check source
- **Instructions**: Contact dataset creators for access

## Using the Download Scripts

### GitHub Downloader (No Account Required)
```bash
python download_from_github.py
```
This script can download datasets directly from GitHub without requiring authentication.

### Kaggle Downloader (Requires Account)
```bash
python download_helper.py
```
or
```bash
download_kaggle_datasets.bat
```

## Quick Start - Download COVID-19 Dataset (Easiest)

The COVID-19 dataset is the easiest to download as it doesn't require authentication:

```bash
# Option 1: Using Python script
python download_from_github.py
# Select option 1

# Option 2: Direct download
git clone https://github.com/ieee8023/covid-chestxray-dataset.git
# Or download ZIP from: https://github.com/ieee8023/covid-chestxray-dataset/archive/refs/heads/master.zip
```

## Disclaimer

These datasets are provided for educational and research purposes only.
Always consult with medical professionals for actual medical diagnosis.