P026
Lung ultrasound for the detection of pulmonary tuberculosis using expert- and AI-guided interpretation: a prospective cohort study
V Suttels(1,2) T Brokowski(3) A P Wachinou(4) J Wolleb(3) A R Hada(4) J D Du Toit(5) G Makpemikpa(4) C Bessat(1) E Garcia(1) A Roux(1) T Brahier(1) O Opota(6) D Affolabi(7) M A Hartley(2) N Boillat-Blanco(1)
1: Lausanne University Hospital; 2: Intelligent Global Health Research Group, Swiss Institute of Technology (EPFL); 3: Yale school of medicine, Department of Biomedical Informatics & Data Science; 4: National Teaching Hospital for Tuberculosis and Pulmonary Diseases (CNHU-PPC); 5: MRC Wits Rural Public Health and Health Transitions Research Unit (Agincourt), Faculty of Health Sciences, University of the Witwatersrand; 6: Institute of Microbiology University of Lausanne and University Hospital Centre, Lausanne; 7: Laboratoire de référence des mycobactéries (LRM), Cotonou, Benin
Background
Point-of-care lung ultrasound (LUS) is a promising tool for portable sputum-free tuberculosis (TB) triage. We investigate the diagnostic performance of LUS to detect TB using expert and artificial intelligence (AI) guided interpretation. We introduce ULTR-AI (Ultrasound-led TB recognition using AI), a suite of deep learning (DL) models designed to automate TB risk stratification from LUS images.
Methods
In this prospective cohort study in a tertiary center in urban Benin, a standardized 14-point sliding scan LUS protocol was performed for symptomatic patients by a trained operator. LUS images were reviewed by two blinded and independent readers. Same-day single sputum Xpert MTB/RIF Ultra® was the microbiological reference standard. The suite comprises three AI models, ULTR-AI, ULTR-AI[signs] and ULTR-AI[max]. ULTR-AI predicts TB directly from images using DL, ULTR-AI[signs] first generates human-recognizable pathological signs before TB risk prediction in a machine learning (ML) model. Finally, ULTR-AI[max] takes the maximal TB risk score predicted by these two models.
Results
Out of 760 screened, 504 were analyzed. Of these, 192 (38%) had bacteriologically confirmed TB. ULTRA-AI[max] reached the recommended WHO requirements for a sputum-free TB triage test of 0.9 sensitivity and 0.7 specificity, achieving a sensitivity of 0.91 (95% CI, 0.90–0.96) and a specificity of 0.85 (95% CI, 0.74–0.88) [AUROC 0.93, 95% 0.92-0.95].
Conclusion
In this cohort, AI-guided lung ultrasound meets the WHO requirements for sputum-free TB triage, enabling point-of-care testing with minimal infrastructure. ULTR-AI could help decentralize TB diagnostics in LMICs, enhancing timely detection. Validation in diverse populations is crucial to confirm clinical utility.