Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening

2026-08-26

AI Summary

研究通过跨数据集评估机器学习和深度学习的咳嗽模型在结核病筛查中的泛化能力，发现模型受数据采集设备影响大，泛化性能不佳。

Abstract

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.

Problem

• ML-based cough models
• cross-dataset generalizability
• tuberculosis screening
• audio representations
• data collection artifacts

Innovation

• cross-dataset generalizability
• cough acoustics
• device diversity
• clinical-variable baseline