Welcome to the Stop TB Partnership and FIND resource centre on computer-aided detection products for the diagnosis of tuberculosis
In recent years, the tuberculosis (TB) field has welcomed several computer-aided detection (CAD) products that provide an automated and standardized interpretation of digital chest X-rays based on artificial intelligence. The AI4HLTH resource centre from the Stop TB Partnership and FIND aims to list all CAD products that are available for TB detection, irrespective of their stage of development, regulatory or policy approval to provide implementors and researchers with relevant information on CAD products.
​
The AI/CAD tools listed on this website have self-declared their regulatory status and performance claims. However, regulatory approvals and real-world performance can vary significantly between vendors. Inclusion in this list does not imply endorsement, validation, or compliance with global procurement standards.
​
In the WHO guidelines on systematic screening (2021), three (3) AI products were reviewed for the screening and triage of TB. They are CAD4TB V6 (Delft Imaging), Lunit INSIGHT CXR V1 (Lunit) and qXR V2 (Qure.ai). WHO is currently reviewing new software versions and additional products, and this page will be updated when new products are included in WHO guidelines.
​
Although several products can be used in children, it is important to note that currently WHO has not recommended any CAD product for individuals under 15. Independent data on performance of CAD in children is an important research gap. In addition, no CAD product has been recommended by WHO for findings beyond TB, although several providers report on findings beyond TB scores. Independent evaluations of non-TB findings are also important to improve the evidence base.
​
This site attempts to document and display regulatory approvals. However, it is important to note the differences in regulatory classifications and approvals from non-stringent in-country regulators as they reflect the depth of testing, post-market surveillance, as well as the ability to support end users.
​
End-users are encouraged to verify regulatory approvals, read independent evaluations, and assess long-term support capabilities before selecting an AI solution for deployment.
Learn more about the features of available and upcoming CAD products for TB in the Products section.
CAD products are developed through artificial intelligence methods using large amounts of data. Upon scanning of a new chest X-ray they provide an abnormality score that can be used to trigger further diagnostic examination in individuals with a score above a certain threshold. CAD products may serve as screening or prioritization tools to facilitate the diagnostic process. A few CAD products are currently on the market and several others are in late stages of development.
​
​In 2020, the Stop TB Partnership and FIND sent out the first survey to developers known to have a market-ready solution or to be working on a CAD product that could be used by TB implementers to interpret chest X-rays to collect information about their product. More details about the approached candidates be found in this publication. Since 2020, updates are made periodically.
​​
Quick links:
-
World Health Organization – Consolidated guidelines on tuberculosis Module 2: Screening – Systematic screening for tuberculosis disease
-
World Health Organization – Operational handbook on tuberculosis Module 2: Screening – Systematic screening for tuberculosis disease
-
FIND – Digital chest radiography and computer-aided detection (CAD) solutions for TB diagnostics
-
Stop TB Partnership – Screening and triage for TB using Computer-Aided Detection (CAD) Technology and Ultra-portable X-Ray Systems: A Practical Guide
Updating the resource center ​​
For developers: Please contact us if your company has a TB AI product that is not yet listed, or if you would like to report an update or error.
DISCLAIMER: The Information presented was supplied directly from the companies and should be considered “as is” (“Information”). Under no circumstances shall Stop TB Partnership or FIND be responsible or warrant for the accuracy of any of the Information presented. This is not an endorsement from Stop TB Partnership and FIND. Stop TB Partnership and FIND shall not be responsible and will not indemnify any third party in any way, for loss, claim, damage, or liability, of whatsoever kind or nature, which may arise from or in connection with the use of the Information. FIND and Stop TB Partnership encourage product developers to provide updates relevant to their products using the "contact us" functionality above to ensure the most up-to-date Information is always provided. Product(s) will be removed from the website if no confirmation on the product status is received from the company for two years.