← 研踪 PaperTrailarXiv:2608.25846 · v2
Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening
首次提交 2026-08-26 22:22
版本更新 2026-11-18 00:04
雷达观测 2026-11-18 16:38
摘要
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 ), both pipelines fail to generalize, with external performance frequently below 0.6, indicating a possible limitation of the data. We further observed that 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 ), indicating acquisition-specific variability in acoustics is a stronger driver of poor generalizability. High within-dataset performance is not enough. External validation is essential before cough-based TB models are clinically ready.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-18 00:04 · 最新观测版本
Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening
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 ), both pipelines fail to generalize, with external performance frequently below 0.6, indicating a possible limitation of the data. We further observed that 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 ), indicating acquisition-specific variability in acoustics is a stronger driver of poor generalizability. High within-dataset performance is not enough. External validation is essential before cough-based TB models are clinically ready.
查看此版本解读 →查看此版本原文 ↗