CAD4TB triage of 94,694 people in Nigeria yields 2,337 TB cases

AI system

System: CAD4TB computer-aided detection on ultra-portable digital chest X-ray
Model: Not assessed
Release: CAD4TB version 7.
Application developer: Delft Imaging
Role: CAD4TB scored chest X-rays for signs of tuberculosis.
Configuration: The deployment used CAD4TB version 7 with a threshold score of 50.
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: Ultra-portable X-ray units with CAD4TB v7 at threshold 50. Community mobilization and same-day Xpert testing. USAID funding through the iNTP project.
Human oversight: Not assessed
Relation to AI: A parallel algorithm classified anyone positive on symptom screening or with a CAD4TB score of 50 or above as presumptive TB. They went for Xpert testing and/or clinical assessment, so the AI score was one of two gates to testing, not the sole gate. Of the 2,337 diagnosed, 1,614 (69.1%) scored 50 or above and 723 (30.9%) scored below 50. The TB yield also depends on X-ray reach, Xpert availability and treatment linkage. Both papers are authored by implementing organizations (KNCV Nigeria, IHVN, USAID, NTBLCP), not the vendor.
AI contribution: A parallel algorithm classified anyone positive on symptom screening or with a CAD4TB score of 50 or above as presumptive TB. They went for Xpert testing and/or clinical assessment, so the AI score was one of two gates to testing, not the sole gate. Of the 2,337 diagnosed, 1,614 (69.1%) scored 50 or above and 723 (30.9%) scored below 50. The TB yield also depends on X-ray reach, Xpert availability and treatment linkage. Both papers are authored by implementing organizations (KNCV Nigeria, IHVN, USAID, NTBLCP), not the vendor.

Benefits

Outcome identifiers: Not assessed
Benefit: 2,337 TB cases diagnosed, 97% treated
Category: Improved Physical Health
Status: Not assessed
Beneficiary groups: People. Organizations. Society.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Measurement

Studies: Not assessed
Study design: Not assessed
Measurement: Not assessed
Setting and sample: Not assessed
Measures: 2,337 TB cases diagnosed, 97% treated
Comparison: KNCV Nigeria's other 2022 case-finding strategies in the same states without portable X-ray (intensified case finding in public and private facilities, medicine vendors and community pharmacists, contact investigation, community active case finding)
Results: Reference: NNS 205.6 (public facilities), 179.0 (private facilities), 127.8 (medicine vendors and community pharmacists), 49.0 (contact investigation), 388.5 (community active case-finding). Result: NNS 44.9 with PDX-CAD.
Additional value from AI: Not assessed
Limitations: Not assessed
Denominators: Not assessed
Materials for replication: Not assessed

Money

Kind of value: Not assessed
Amount: Not assessed
Currency and prices: Not assessed
Period and unit: Not assessed
Realization: Not assessed
Recipients and payers: Not assessed
Gross or net: Not assessed
Included costs: Not assessed
Comparison: Not assessed
Method: Not assessed
Return and payback: Not assessed

People and places

Participants: Developer: Delft Imaging. Operator: KNCV Tuberculosis Foundation Nigeria and Institute of Human Virology Nigeria with the National TB and Leprosy Control Program, funded by USAID.
Jurisdictions: Benefit: Eight states, Nigeria. Developer: Netherlands. Operator: Nigeria.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Follow-up

Conditions for replication: Threshold calibrated to local prevalence. Xpert cartridge supply and sample transport. Treatment linkage within days.
Replication attempts: 2 follow-up entries. Benefit recurred: 1. Outcome unresolved: 1.
Later findings: 30 September 2025. KNCV Nigeria's 2025 annual report says the IMPACT project completed deployment of 390 portable digital X-ray units with AI across all 36 states and the FCT, adding 266 units in 2025, and screened over 780,000 people with over 23,000 TB cases diagnosed in January to September 2025.
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Evidence

Source numbers: AIBD-0096-source-0001–0003
Citations: Numbered references below.
Language: Not assessed
Source relationships: Operator
Original evidence: Not assessed
Access and versions: Not assessed
Supported findings: Not assessed
Evidence links: Not assessed
Evidence strength: Not assessed
Missing evidence: Not assessed

Grades

Assessed benefits: People. Organizations. Society.
Depth: Not assessed
Reasons: People: Aggregate screening yield; the sources do not document any individual's outcome attributable to the software. Organizations: Efficiency figures (NNS 38, NNT 4) are program aggregates without a discrete attributable event. Society: 2,337 diagnoses are an aggregate program result.
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: The deployer's account

Entry details

Kind

Not assessed

Related entries

Chest X-ray classification systems, the same kind as CAD4TB, underdiagnosed economically vulnerable, female and Hispanic patients because of biased training data.

Field

Global health: tuberculosis case finding. Public health program.

Purpose

Triage chest X-rays to select presumptive TB cases for molecular testing during community active case finding.

Dates

Date basis unassessed: 2022 (precision unassessed). Updated: 21 October 2026.

Stage

Not assessed

Version

SHA-256: ceffb7ad68c5…

References

Other identifiers: AI Incident Database: 81.
Category mappings: Not assessed
Brief: Not assessed
Revisions: 7 revisions. Latest: 21 October 2026.

Sources

Peer-reviewed study Deployer

Alege, A., Useni, S., Ihesie, A., Eneogu, R., Agbaje, A., et al. (2025). Impact of the use of the ultra-portable digital x-ray with CAD4TB for active case finding for tuberculosis in Nigeria. Frontiers in Digital Health.
Peer-reviewed study Deployer

Useni, S., Odume, B., Tukur, M., et al. (2025). Comparative analysis of the impact of portable digital X-ray on TB screening in hard-to-reach areas in Nigeria. Public Health Action 15(1), 17–20.
Report Deployer

KNCV Nigeria (2026). 2025 Annual Report. Knowledge Network for disease Control and Vigilance, Abuja (PDF posted March 2026; day not given, first day of month used).