Search papers, labs, and topics across Lattice.
This study systematically evaluates the generalizability of machine learning (ML) models for tuberculosis (TB) screening using cough acoustics across three independent datasets. Despite achieving moderate performance within individual datasets (ROC-AUC up to 0.755), the models consistently underperform in external validation, often falling below an ROC-AUC of 0.6, suggesting that data collection artifacts significantly impact model effectiveness. The findings highlight that audio representations are more influenced by recording devices and datasets than by actual TB status, emphasizing the need for external validation before clinical deployment of cough-based TB screening models.
High-performing cough-based TB models fail to generalize across datasets, revealing that data collection artifacts overshadow disease-related signals.
Cough acoustics are promising for non-invasive tuberculosis (TB) screening, yet whether machine learning (ML) models capture disease-related acoustics or artifacts of data collection remains unresolved. We evaluated the cross-dataset generalizability of classical ML and deep learning (DL) cough-based TB classifiers across three independent datasets. Despite moderate within-dataset performance (ROC-AUC up to $0.755 \pm 0.056$), both pipelines fail to generalize, with external performance frequently below 0.6, indicating a possible limitation of the data. We further observed audio representations are organized by recording device and dataset rather than TB status, predicted TB probability tracks country-level prevalence in CODA, and device mismatch degrades transfer while device-diverse training improves it. Additionally, a clinical-variable baseline generalizes more consistently (ROC-AUC $0.655 - 0.711$), indicating acquisition-specific variability is a stronger driver of poor generalizability than population shift. High within-dataset performance is not enough. External validation is essential before cough-based TB models are clinically ready.
