TB/COVID screening

Ministries of Health and funders like Global Fund and USAID increasingly focus on bi-directional screening and testing for TB and COVID. Digital X-ray and Artificial Intelligence support these screening demands in resource-constrained settings with innovative and proven CAD systems to simultaneously detect and quantify abnormalities suggestive of TB and COVID within less than 1 minute.

Funders call for bi-directional screening of TB & COVID

The Global Fund

“We are providing guidance, tools and immediate funding of more than US$1 billion to help countries fight COVID-19, mitigate the impacts on lifesaving HIV, TB and malaria programs, and prevent fragile health systems from being overwhelmed.”
COVID-19 - The Global Fund to Fight AIDS, Tuberculosis and Malaria

  • The COVID-19 pandemic is having a catastrophic impact on the most vulnerable communities around the world and threatens progress against HIV, TB and malaria.
  • Investments in global health are critical to keep communities safe from infectious diseases including HIV, TB and malaria, and to help fight new health threats like COVID-19.
  • Diseases know no borders and the COVID-19 pandemic makes clear that our global health security is only as strong as the world’s weakest health system.”

Stop TB Partnership

  • Finding the missing case with TB that increased to ~4.5 million due to COVID.
  • Bi-directional screening of TB and COVID-19 for people displaying influenza-like illness and severe acute respiratory infections. Contact tracing systems and testing for TB linked to COVID-19 contact tracing.
    • Integrated screening and testing for respiratory symptoms; massive community screening & testing for both TB & COVID-19
    • Invest in Technology: X-ray services, AI and rapid molecular tests; scale is critical
  • In addition to the worldwide drop in TB diagnosis and treatment, data emerging from India and South Africa shows that people coinfected with TB and COVID-19 have three times higher mortality than people infected with TB alone. This makes contact tracing, case finding and bi-directional TB and COVID-19 testing essential.


  • TB and COVID-19 symptoms may overlap, and patients may be co-infected
  • COVID-19 and TB screening algorithms and evaluation pathways should be bi-directional.
  • All HIV positive patients should be screened for both TB and COVID.

Artificial Intelligence for COVID control

The fight against COVID-19 inspired specialists in various countries to unite and collaborate to develop Artificial Intelligence software to more effectively detect COVID-19. The Dutch AI innovator Delft Imaging initiated the development of CAD4COVID at the wake of the pandemic in cooperation with Radboud University, Thirona and COVID hot spot hospitals in Europe and Mexico. Support was also provided through the Dutch Government RVO Agency and Dutch FMO Development Bank. As a social enterprise, Delft Imaging decided to make CAD4COVID available free-of-charge for both CT and X-ray COVID-19 screening. This CE Certified Artificial Intelligence software provides hospitals in over 25 countries with pro bono access to an e-Radiologist at Dutch Senior Radiologist accuracy level for rapid detection and quantification of COVID-19 signs in lungs. CAD4COVID comprises two different algorithms: CAD4COVID-XRay and CAD4COVID-CT.

CAD4COVID could be effectively combined in resource constrained settings with an oxygen-meter and the C-Reactive Protein (CRP) blood test using a portable Hemocue device. The c-reactive protein test measures the level of c-reactive protein (CRP) in the blood that responds to inflammation. First test results in Southern Africa indicate that the CRP score (normally < 4 for healthy individuals and often > 100 for COVID-19 patients) in combination with CAD4COVID abnormality score increases diagnostic accuracy and supports better informed clinical pathways especially in those settings where PCR tests are not available or require substantial time to process.

Lungs' CAD image

As a next step in response to the public health demands in Ghana the enhanced performance of the latest released CAD4TB version is now seamlessly integrated in the CAD4TB/COVID box for bi-directional rapid screening nationwide at more than 50 sites. The user can select CAD4TB only and opt for bi-directional screening for COVID and TB as long as required during the pandemic. The use of CAD4TB is payable per use and its cost depends on the image bundle size.

CAD4COVID-XRay how it works

CAD4COVID can be used for chest X-ray interpretation next to symptom screening as a first-line rapid triage for COVID-19 and disease severity indicator, before any further testing. CAD4COVID-XRay calculates within 30 seconds an abnormality severity score between 0 and 100 indicating the extent of presumptive COVID-19 related abnormalities. In addition, the AI highlights lung abnormalities through a multi-color heat-map. The software can be used in the cloud by uploading the images or on-site / off-line with a compact box.

According to the Radiology Journal publication, the CAD4COVID-XRay AI system performed at the same level as independent expert readers for the detection of COVID-19 characteristics.
Read the study published in the Radiology journal.

CAD4COVID-XRay can be used – after symptom screening – to:

  1. automatically detect and indicate in the heat-map the location of abnormalities in the chest radiograph suggestive of COVID-19
  2. assess the COVID-19 probability score (0 – 100) to allow care providers taking more informed decisions on e.g. testing, self-quarantine or hospitalization in specialized ward for care before the outcome of the confirmatory test (RT-PCR) is received, resulting in reduced transmission of the virus and protection of health care providers
  3. quantify the % of visible affected lung tissue to track disease severity and patient recovery

Ghana an AI front-runner to control COVID

The compact AI box used in Ghana hosts an e-Radiologist that proved to be as accurate in detecting presumptive COVID-19 abnormalities as senior radiologists in the Netherlands, within less than 1 minute. CAD4COVID is another digital innovation in support of universal health coverage and equity in public health.


With the objective to facilitate rapid triaging for COVID-19 Ghana Health Service, an early adopter of eHealth innovations, launched as the first country in Africa the nationwide use of CAD4COVID-XRay from July 2020. This AI supports better clinical decisions on self-isolation or hospitalization of patients at initially about 50 public hospitals across Ghana and is in support of the national strategy to digitize diagnostic imaging to make the health system more resilient. In order to support the fight against Tuberculosis - the same AI box now also provides a heat-map and probability score for the presence in digital chest X-rays of abnormalities suggestive of TB.

CAD4TB/COVID-XRay is the first evidence based and CE certified algorithm to facilitate accurate and rapid bi-directional screening for TB and COVID in resource constrained settings.