The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.

🫁 Tuberculosis Chest X-Ray RAG Database (Part 3)

A specialized Retrieval-Augmented Generation (RAG) database for tuberculosis detection from chest X-ray images, built using DenseNet-121 deep learning model and FAISS vector search.

Developed by: Gaston Software Solutions LLP
Location: Ntinda, Kampala, Uganda
Contact: info@gss-tec.com
Website: www.gss-tec.com

📊 Database Overview

Statistics

Total Images: 4,200 chest X-ray images
Embedding Dimension: 1,024 features per image
Model: DenseNet-121 (pretrained on ImageNet)
Vector Index: FAISS (GPU-accelerated)
Dataset: Tuberculosis TB Chest X-Ray Dataset

Dataset Distribution

Class	Images	Percentage	Description
Normal	3,500	83.3%	Healthy chest X-rays
Tuberculosis	700	16.7%	TB-positive chest X-rays
Total	4,200	100%	TB screening dataset

Class Imbalance Note

This dataset reflects real-world TB screening scenarios where:

Normal cases are more common (83.3%)
TB cases are less frequent (16.7%)
Ratio approximates actual TB prevalence in screening programs

🔬 Tuberculosis Classes

1. Normal (3,500 images)

Type: Healthy chest X-rays
Characteristics:
- Clear lung fields
- Normal cardiac silhouette
- No infiltrates or consolidation
- No cavitation or nodules
Purpose: Baseline for TB detection
Clinical Use: Negative screening results

2. Tuberculosis (700 images)

Type: TB-positive chest X-rays
Characteristics:
- Pulmonary infiltrates
- Cavitary lesions
- Nodular opacities
- Upper lobe predominance
- Possible pleural effusion
Severity: Active pulmonary tuberculosis
Clinical Use: Positive screening requiring treatment

📁 Database Files

The RAG database consists of three essential files:

1. `embeddings.npy` (~16 MB)

Format: NumPy array
Shape: (4200, 1024)
Content: DenseNet-121 feature vectors for all chest X-rays
Purpose: Dense vector representations for similarity search

2. `metadata.json` (~1.1 MB)

Format: JSON
Content: Image metadata including:
- image_path: Original file path
- class: TB classification (Normal or Tuberculosis)
- dataset: Source dataset name (tuberculosis)
- filename: Image filename
Purpose: Linking search results back to original X-rays

3. `faiss_index.bin` (~16 MB)

Format: FAISS binary index
Type: IndexFlatIP (Inner Product)
Content: Optimized vector search index
Purpose: Fast similarity search across 4,200 chest X-rays

🚀 Usage

Loading the Database

import numpy as np
import faiss
import json
from pathlib import Path

# Load embeddings
embeddings = np.load('embeddings.npy')

# Load metadata
with open('metadata.json', 'r') as f:
    metadata = json.load(f)

# Load FAISS index
index = faiss.read_index('faiss_index.bin')

print(f"Loaded {index.ntotal} chest X-ray vectors")

# Check class distribution
tb_count = sum(1 for m in metadata if 'Tuberculosis' in m['class'])
normal_count = sum(1 for m in metadata if 'Normal' in m['class'])
print(f"TB cases: {tb_count}, Normal cases: {normal_count}")

Performing Similarity Search

from PIL import Image
import torchvision.transforms as transforms
import torchvision.models as models
import torch

# Load DenseNet-121 model
model = models.densenet121(weights=models.DenseNet121_Weights.IMAGENET1K_V1)
model.classifier = torch.nn.Identity()
model.eval()

# Image preprocessing
transform = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], 
                       std=[0.229, 0.224, 0.225])
])

# Extract features from query chest X-ray
def extract_features(image_path):
    img = Image.open(image_path).convert('RGB')
    img_tensor = transform(img).unsqueeze(0)
    
    with torch.no_grad():
        features = model(img_tensor).numpy()
        # Normalize
        features = features / np.linalg.norm(features)
    return features

# Search for similar chest X-rays
query_features = extract_features('patient_xray.jpg')
k = 10  # Number of similar X-rays to retrieve

distances, indices = index.search(query_features, k)

# Get results
print("Similar chest X-rays:")
for i, idx in enumerate(indices[0]):
    result = metadata[idx]
    diagnosis = result['class'].replace('tuberculosis_', '')
    print(f"{i+1}. {diagnosis.upper()} - {result['filename']}")
    print(f"   Similarity: {distances[0][i]:.4f}")

TB-Specific Search

def search_tb_cases(query_features, k=5):
    """Search only among TB-positive cases"""
    # Get indices of TB cases
    tb_indices = [i for i, m in enumerate(metadata) 
                  if 'Tuberculosis' in m['class']]
    
    # Create TB-only index
    tb_embeddings = embeddings[tb_indices]
    tb_index = faiss.IndexFlatIP(1024)
    faiss.normalize_L2(tb_embeddings)
    tb_index.add(tb_embeddings)
    
    # Search
    distances, indices = tb_index.search(query_features, k)
    
    # Map back to original indices
    results = []
    for idx in indices[0]:
        original_idx = tb_indices[idx]
        results.append(metadata[original_idx])
    
    return results, distances[0]

# Find similar TB cases
tb_results, scores = search_tb_cases(query_features, k=5)
print("\nSimilar TB cases:")
for i, result in enumerate(tb_results):
    print(f"{i+1}. {result['filename']} - Score: {scores[i]:.4f}")

Screening Support System

def tb_screening_support(query_features, threshold=0.75):
    """
    Provide screening support by finding similar cases
    Returns: (similar_cases, tb_probability_estimate)
    """
    k = 20  # Check top 20 similar cases
    distances, indices = index.search(query_features, k)
    
    # Count TB cases in top results
    tb_count = 0
    similar_cases = []
    
    for i, idx in enumerate(indices[0]):
        if distances[0][i] >= threshold:  # Only high similarity
            result = metadata[idx]
            similar_cases.append({
                'filename': result['filename'],
                'class': result['class'],
                'similarity': float(distances[0][i])
            })
            if 'Tuberculosis' in result['class']:
                tb_count += 1
    
    # Estimate TB probability based on similar cases
    tb_probability = tb_count / len(similar_cases) if similar_cases else 0
    
    return similar_cases, tb_probability

# Use for screening
similar, tb_prob = tb_screening_support(query_features)
print(f"\nTB Probability Estimate: {tb_prob*100:.1f}%")
print(f"Based on {len(similar)} similar cases")

🧠 Model Architecture

DenseNet-121

Architecture: Dense Convolutional Network
Layers: 121 layers with dense connections
Feature Dimension: 1,024
Pretrained: ImageNet-1K
TB Detection Advantages:
- Excellent feature extraction for chest X-rays
- Captures subtle pulmonary patterns
- Proven performance on medical imaging
- Efficient for large-scale screening

📈 Performance Characteristics

Search Performance

Index Type: Flat (exact search)
Distance Metric: Inner Product (cosine similarity)
Search Speed: <1ms per query (GPU)
Accuracy: 100% (exact nearest neighbors)

Memory Requirements

Total Size: ~33 MB (all three files)
RAM for Loading: ~60 MB
GPU Memory: ~500 MB (for model inference)

Dataset Characteristics

Class Imbalance: 5:1 ratio (Normal:TB)
Real-world Reflection: Mirrors actual screening scenarios
Considerations:
- More normal cases for robust baseline
- Sufficient TB cases for pattern learning
- Balanced for screening applications

🔧 Technical Details

Feature Extraction Pipeline

Input: Chest X-ray images (grayscale or RGB)
Preprocessing:
- Convert to RGB if grayscale
- Resize to 256×256
- Center crop to 224×224
- Normalize with ImageNet statistics
Model: DenseNet-121 (without classification layer)
Output: 1,024-dimensional feature vector
Normalization: L2 normalization for cosine similarity

Vector Index

Algorithm: FAISS IndexFlatIP
Normalization: L2-normalized vectors
Similarity: Cosine similarity via inner product
GPU Support: Yes (for faster search)

📚 Use Cases

1. TB Screening Support

Assist radiologists in TB detection
Find similar historical TB cases
Support differential diagnosis
Reduce false negatives in screening

2. Quality Assurance

Second opinion for TB diagnosis
Peer review of screening results
Training validation for radiologists
Consistency checks across readers

3. Medical Education

Teaching TB radiological patterns
Case-based learning
Demonstrate disease progression
Train medical students and residents

4. Research Applications

Study TB presentation patterns
Analyze radiological features
Build training datasets for AI
Validate diagnostic algorithms

5. Public Health Programs

Mass screening campaigns
Contact tracing support
Treatment monitoring
Epidemiological studies

🌍 Global TB Context

Tuberculosis Facts

Global Burden: ~10 million new TB cases annually
Mortality: One of top 10 causes of death worldwide
Screening: Chest X-ray is primary screening tool
Challenge: Requires expert radiologist interpretation
Solution: AI-assisted screening can improve access

Screening Importance

Early Detection: Critical for treatment success
Contact Tracing: Identify exposed individuals
Public Health: Control TB transmission
Resource-Limited: AI support in areas with few radiologists

🔒 Data Privacy & Ethics

Important Considerations

Public Dataset: From publicly available Kaggle dataset
De-identification: Ensure patient privacy in deployment
Clinical Use: Screening support tool, not diagnostic device
Validation: Always confirm with qualified radiologists
Regulatory: Requires medical device approval for clinical use
Bias: Be aware of dataset limitations
False Negatives: Critical to minimize in TB screening

Clinical Deployment Guidelines

Not a Replacement: Supplement, not replace, radiologist review
Validation Required: Extensive clinical validation needed
Regulatory Approval: Obtain necessary certifications
Quality Control: Regular performance monitoring
Training: Proper training for healthcare workers
Documentation: Clear limitations and use cases

🛠️ Building the Database

The database was built using:

Notebook: Health_Imaging_RAG_Colab_Part3.ipynb
Platform: Google Colab with T4 GPU
Processing Time: ~8-12 minutes
Framework: PyTorch + FAISS
Model: DenseNet-121 (torchvision)

Dataset Source

Kaggle: tawsifurrahman/tuberculosis-tb-chest-xray-dataset
Original Size: 4,200 chest X-rays
Format: JPEG/PNG images
Classes: 2 classes (Normal, Tuberculosis)

📊 Dataset Statistics

Image Distribution

Normal:        ████████████████████████████████████████ 3,500 (83.3%)
Tuberculosis:  ████████                                   700 (16.7%)

Quality Metrics

Imbalance Ratio: 5:1 (reflects real-world screening)
Image Quality: Clinical-grade chest X-rays
Coverage: Diverse TB presentations
Reliability: Suitable for screening support systems

📖 Citation

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

@software{tuberculosis_rag_2026,
  author = {{Gaston Software Solutions LLP}},
  title = {Tuberculosis Chest X-Ray RAG Database},
  year = {2026},
  publisher = {GSS-LLP},
  address = {Ntinda, Kampala, Uganda},
  url = {https://www.gss-tec.com},
  note = {TB screening support system using DenseNet-121 and FAISS}
}

Please also cite the original dataset:

@dataset{tuberculosis_xray,
  author = {Tawsifur Rahman},
  title = {Tuberculosis (TB) Chest X-ray Database},
  year = {2020},
  publisher = {Kaggle},
  url = {https://www.kaggle.com/datasets/tawsifurrahman/tuberculosis-tb-chest-xray-dataset}
}

📞 Contact & Support

Gaston Software Solutions LLP

Email: info@gss-tec.com
Website: www.gss-tec.com
Location: Ntinda, Kampala, Uganda

For technical support, questions, or collaboration opportunities, please contact us via email.

⚖️ License

MIT License

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Intellectual Property Notice

This Tuberculosis Chest X-Ray RAG Database is provided by Gaston Software Solutions LLP (GSS-LLP) under the MIT License for open-source use.

While the software is open-source and freely available for use, modification, and distribution under the MIT License terms, the intellectual property rights, including the database architecture, processing pipeline, and implementation methodology, remain the property of Gaston Software Solutions LLP.

Key Points:

✅ Free to use, modify, and distribute under MIT License
✅ Open-source for research and commercial applications
✅ No warranty or liability from GSS-LLP
⚠️ Intellectual property and methodology rights retained by GSS-LLP
⚠️ Original dataset subject to its respective license
⚠️ Medical use requires appropriate validation and regulatory compliance
⚠️ Not approved for clinical diagnosis without proper medical device certification
⚠️ TB screening requires expert radiologist confirmation

Dataset License

The underlying tuberculosis chest X-ray dataset is sourced from Kaggle and is subject to its respective license. Users must comply with the original dataset license when using this RAG database. Please refer to:

Tuberculosis TB Chest X-Ray Dataset: https://www.kaggle.com/datasets/tawsifurrahman/tuberculosis-tb-chest-xray-dataset

🌟 About Gaston Software Solutions LLP

Gaston Software Solutions LLP (GSS-LLP) is a technology company based in Ntinda, Kampala, Uganda, specializing in AI/ML solutions, software development, and data science applications for healthcare and other industries.

Our Mission: Leveraging cutting-edge technology to solve real-world problems and improve lives through innovative software solutions.

Healthcare AI Services:

TB Screening Support Systems
Medical Image Analysis
RAG Databases for Healthcare
Clinical Decision Support Tools
Public Health AI Solutions

Visit us at www.gss-tec.com or contact info@gss-tec.com for more information.

🔗 Related Databases

This is Part 3 of the Health Imaging RAG Database series:

Part 1: Pneumonia + Malaria (66,828 images)
Part 2: Brain Tumor MRI (7,200 images)
Part 3: Tuberculosis (4,200 images) ← You are here

Combined Total: 78,228 medical images across multiple conditions

Last Updated: June 2026
Version: 1.0
Total Images: 4,200
Model: DenseNet-121
Developed by: Gaston Software Solutions LLP
License: MIT

Downloads last month: 28

Total file size:

35.4 MB