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 $0.755 \pm 0.056$), both pipelines fail to generalize, with external performance frequently below 0.6, indicating a possible limitation of the data. We further observed audio representation
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 52%Multi-Source Data-Driven Machine Learning Reshapes the Diagnosis and Treatment of Lung Cancer - EIN Presswire →
- PossiblePossibly related (embedding) · 50%Multi-source data-driven machine learning improves lung cancer care - News-Medical →
- PossiblePossibly related (embedding) · 49%Code-Free Classification of Pediatric Pneumonia on Chest Radiographs Using Google Cloud Vertex AI AutoML: A Proof-of-Concept Internal Validation Study - Cureus →
- PossiblePossibly related (embedding) · 51%Machine learning predicts tuberculosis drug resistance from whole genomes, review finds - Bioengineer.org →
- LinkedLinked via arxiv author · 85%Wensi Zhang →
“Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening”
- LinkedLinked via arxiv author · 85%Tomas Teijeiro →
“Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening”
- LinkedLinked via arxiv author · 85%Jérôme Thevenot →
“Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening”
- LinkedLinked via arxiv author · 85%David Atienza →
“Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening”