←Back to feed
🧠 AI🟢 BullishImportance 6/10
Predicting Tuberculosis from Real-World Cough Audio Recordings and Metadata
arXiv – CS AI|George P. Kafentzis, Stephane Tetsing, Joe Brew, Lola Jover, Mindaugas Galvosas, Carlos Chaccour, Peter M. Small||2 views
🤖AI Summary
Researchers developed an AI system that can detect tuberculosis from cough recordings with 70% accuracy using audio alone, improving to 81% when combined with clinical metadata. The study used real-world data from a phone-based app across Africa and Asia, suggesting mobile applications could enhance TB diagnosis in community health settings.
Key Takeaways
- →AI achieved 70% accuracy (AUC 0.70) in detecting tuberculosis from cough audio recordings alone
- →Accuracy improved to 81% when combining cough analysis with demographic and clinical factors
- →Study used large dataset from real-world phone app recordings across Africa and Asia without manual annotation
- →Mobile phone-based TB screening could reduce costs and improve case-finding for community health workers
- →Research demonstrates potential for automated respiratory disease diagnosis through audio analysis
#ai-healthcare#machine-learning#medical-diagnosis#mobile-health#tuberculosis#audio-analysis#public-health#disease-detection
Read Original →via arXiv – CS AI
Act on this with AI
This article mentions $CRV.
Let your AI agent check your portfolio, get quotes, and propose trades — you review and approve from your device.
Related Articles