Researchers showed AI can detect tuberculosis from mobile photos of analog chest x-rays in low-resource settings.
Key Details
- 1Study used 498 photographed chest x-ray films from Ethiopia and Guinea-Bissau for TB diagnosis comparison.
- 2AI tool (qXR, Qure.ai) identified 81 TB cases vs. 50 and 99 by two radiologists.
- 3AI achieved sensitivity of 76.5% and specificity of 85.9% (AUC 0.84) compared with lab-confirmed cases.
- 4AI performance was comparable to experienced radiologists in detecting TB.
- 5Study highlights mobile-based AI screening in areas with limited digital imaging resources.
- 6Moderate inter-reader agreement (kappa = 0.45–0.56) between radiologists and AI.
Why It Matters
This work suggests mobile phone-based AI can bridge diagnostic gaps in TB screening where digital radiology resources are lacking, expanding access to timely diagnosis and care. If validated, this approach could help advance global health initiatives against tuberculosis, particularly in low-income countries.
Source
AuntMinnie
Related News
•HealthExec
UT Austin to Launch $1B+ AI-Driven Medical Center With Major Gift
UT Austin receives $750 million from Michael and Susan Dell toward a $1 billion+ AI-native medical center opening in 2030.
•AuntMinnie
Radiology Receives Declining Share of Industry Research Funding
Radiologists received only 1.1% of industry-funded research payments in 2024, with a continuing downward trend.
•AuntMinnie
GPT-4o AI Matches Radiologists in Follow-Up Imaging Recommendations
GPT-4o matched the performance of experienced radiologists and surpassed residents in recommending follow-up imaging from routine radiology reports.