Ethiopia flag
Case Study

Ethiopia – Improving TB diagnosis using CAD4TB and Stool Xpert testing

A hand picks the CAD4TB box out of the suitcase

Ethiopia remains among the countries with the higest tuberculosis and TB/HIV burden globally. Early diagnosis is particularly challenging among household contacts who are unable to product sputum. Innovative screening approaches combining CAD4TB and chest radiography with non-respiratory molecular testing may help close these diagnostic gaps. 

A prospective diagnostic accuracy study was conducted among tuberculosis index patients and their household contacts at a referral hospital in Oromia.

Participants underwent digital chest X-ray screening with CAD4TB support and provided stool and sputum samples for Xpert MTB/RIF testing.

Result

CAD4TB demonstrated strong diagnostic performance

77 %
sensivity
>90 %
specificity
  • Performance was higher in adults than in children.
  • Stool Xpert testing showed high agreement with sputum Xpert results, enabling microbiological confirmation among individuals unable to expectorate.
  • Over 10% of household contacts were newly diagnosed with TB.
A hand picks the CAD4TB box out of the suitcase
Conclusion
  • Combining CAD4TB with stool-based molecular testing improves TB detection among household contacts in high-burden, low-resource settings.
  • This integrated approach is particularly valuable for children and others unable to produce sputum, and where access to radiological expertise is limited.

REFERENCE: Guido G. et al (2026). Improving Tuberculosis Diagnosis Through Artificial Intelligence (CAD4TB) and Stool Xpert MTB/RIF Testing: A Prospective Study from Oromia, Ethiopia. Open Forum Infectious Diseases.