Case Study

Nigeria – Detecting asymptomatic TB with Delft Light & CAD4TB

Delft Light used in World TB Day in Nigeria

AI-supported chest X-ray screening enabled the detection of TB among individuals without symptoms in hotspot communities. The intervention demonstrated that CAD4TB-supported screening can identify TB cases that would be missed by symptom-based approaches alone.

Addressing missed TB cases

Intensive TB case finding was implemented in hotspot communities to reduce undetected TB cases.

Screening beyond symptoms

All individuals were screened regardless of symptoms using the WHO four-symptom screen and chest X-ray with CAD4TB.

AI-supported X-ray analysis

Chest X-rays were processed using AI-powered CAD4TB to identify presumptive TB cases.

Confirmatory testing and review

Individuals with CAD4TB scores above 50 were evaluated using GeneXpert, with radiologist review for those unable to produce sputum.

Result

Between January 2022 and January 2024, A total of 25,993 individuals were screened,

25 ,993 people
were screened
6 59 people
screened without TB symptoms
3 9 cases
diagnosed with TB
1 23 (60%) presumptive TB
identified among asymptomatic individuals
  • 11 individuals were diagnosed with TB: 1 was bacteriologically confirmed, and 10 were clinically diagnosed by radiologists.
  • The Number Needed to Screen (NNS) was 60, and the Number Needed to Treat (NNT) was 4.
Delft Light used in World TB Day in Nigeria
Conclusion
  • The use of CXR with AI has proven to be a game changer in diagnosing TB among non-symptomatic individuals in Nigeria.
  • The deployment of more systems targeting high-risk groups is essential to find all TB cases actively.

REFERENCE: Oyawale, M. et al (2024, November 12-16). Closing the TB case finding gap through artificial intelligence (AI)-aided screening of non-symptomatic population: Katsina State experience [Presentation]. The Union World Conference on Lung Health, Bali, Indonesia.