SemanticMD AI for TB

Automatically detect tuberculosis in chest X-rays


Learn More

Background

Tuberculosis (TB) is one of the leading causes of death worldwide. In 2015, more than 10 million people fell ill with TB and 1.8 million died from the disease.

Chest X-rays (CXR) play a crucial role in TB diagnosis, especially in the case of pulmonary TB (PTB), which is one of the most common presentations of TB. Although CXRs do not provide ground truth for confirming TB, they still offer a high sensitivity method for detecting TB-related abnormalities in the lungs (scars, opacities, pleural effusion, etc.). In addition, since CXRs provide a low-cost, rapid examination even in remote settings, it has been recognized as a powerful screening test for TB, especially in areas and populations with higher disease burden.

While the cost of acquiring a CXR has become much more affordable, the interpretation of CXR scans is currently limited by cost and access to trained radiologists. Hence, there are many patients that get diagnosed too late and unable to treat their symptoms using conventional TB antibiotics.

Solution**

We developed AI for TB to improve the quality and efficiency of TB screening programs. Benefits of implementing SemanticMD AI solutions include:

  • AI is fine-tuned to be specific to local patient populations
  • Cost-efficient compared to teleradiology
  • Rapid reading and reporting (<1 second per scan)
  • Useful for screening and triage of patients with multiple symptoms
  • Sensitivity and specificity equivalent to an expert radiologist
  • Can be used offline in remote, low-power areas

**Not available for diagnostic use in the U.S.

Technical Specifications

Artificial intelligence (AI) can immediately analyze digital X-ray images. Our AI software provides a probability score from 1 to 100 for signs of tuberculosis (TB).

Deep learning technologies enable the efficient training and deployment of AI for TB screening. Our algorithm is trained on a growing database of thousands of images from the U.S., India, China, Eastern Europe, and South Africa. Based on this analysis a score for each patient is computed.

The software can run automatically after a digital X-ray is captured. We support DICOM and HL7 standards to route images which enable integration with PACS, EHR and other hospital IT infrastructure. Our software can be deployed in the cloud or run offline.

Product Resources

AI for TB Product Guide

Learn how our AI solutions can help you reduce costs and improve accuracy in your TB screening programs

Download

AI for TB API

Learn how our APIs can help you build, extend and combine SemanticMD AI solutions with your PACS or EHR

Documentation

References

AI for TB has been developed using SemanticMD's proprietary AI platform spun out of Carnegie Mellon University. In an effort to provide the best possible care for patients, we work with the top researchers and academic partners to continually refine our algorithms. Below are some papers from our technical advisor, Dr. Paras Lakhani:

1. Lakhani, Paras, and Baskaran Sundaram. "Deep Learning at Chest Radiography: Automated Classification of Pulmonary Tuberculosis by Using Convolutional Neural Networks." Radiology (2017): 162326.

Partners