Measured, not labeled: un-analyzable input versus misclassification in single-lead wearable ECG
Physiological Measurement · 30.11.2026 · 47 (11) · 115006
DOI: 10.1088/1361-6579/aebae3 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physiological Measurement, online erschienen am 11.11.2026. Verfasst von Yunfei Cheng und 4 weiteren Autor:innen. Beteiligte Einrichtung: Health Technology and Engineering Research Center, SKG Health Technologies Co., Ltd, Shenzhen 518063, Guangdong, People's Republic of China. Hauptthemen laut MEDLINE: Wearable Electronic Devices, Electrocardiography, Signal Processing, Computer-Assisted. Die Arbeit nennt 47 Literaturangaben.
Abstract (Originalsprache)
Objective . Wearable single-lead ECG has expanded arrhythmia screening, but many recordings are inconclusive, and each corpus defines that word differently. We ask how much deployment error is un-analyzable input rather than misclassification when analyzability is measured with one instrument. Approach . We evaluate three diagnostic databases in-domain and leave-dataset-out under a strict loss counting windows carrying each corpus’s inconclusive label as errors unless deferred. We score analyzability with an unsupervised morphological criterion calibrated against three experts on free-living ECG. Reference unreadability is sparse and imbalanced: only seven of 15 subjects contain an unreadable window and one supplies 88% of all unreadable windows. Thresholded sensitivity and every prevalence or true-share correction are therefore exploratory. Main results . The criterion flags almost all expert-unreadable windows but only subsets of the windows labeled inconclusive by their corpora: an inconclusive rate measures convention as much as signal. At τ a = 0.50 , criterion-positive windows are 8.1%, 56.7% and 69.6% of the strict proxy loss on Holter, handheld and patch. The true-share point intervals lie below 50% on handheld and above it on patch, but joint calibration-and-sampling envelopes span 50% on both and still do at the Youden threshold 0.60. On handheld, the calibration-only envelope is more than twice as wide as the corpus-sampling-only envelope. Gate-versus-confidence ordering changes with the accounting and corpus; no general selective-risk advantage of the gate is established. Significance . Corpus labels, criterion flags and latent expert unreadability must not be conflated. A gate can win by predicting a labeling convention, and this retrospective study validates neither transfer to smartwatch dry electrodes nor improvement in patient outcomes. Measure the inconclusive rate with a stated criterion on the raw signal; do not inherit it from a corpus label.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Health Technology and Engineering Research Center, SKG Health Technologies Co., Ltd, Shenzhen 518063, Guangdong, People's Republic of China.
ORCID-Profile: Yunfei Cheng, Peng Deng, Yun Lian, Ruichen Yang, Huibin Sun
Zitieren
APA
Cheng, Y., Deng, P., Lian, Y., Yang, R., & Sun, H. (2026). Measured, not labeled: un-analyzable input versus misclassification in single-lead wearable ECG. Physiological Measurement, 47(11), 115006. https://doi.org/10.1088/1361-6579/aebae3
Vancouver
Cheng Y, Deng P, Lian Y, Yang R, Sun H. Measured, not labeled: un-analyzable input versus misclassification in single-lead wearable ECG. Physiological Measurement. 2026;47(11):115006. doi:10.1088/1361-6579/aebae3
BibTeX
@article{Cheng2026,
author = {Cheng, Yunfei and Deng, Peng and Lian, Yun and Yang, Ruichen and Sun, Huibin},
title = {Measured, not labeled: un-analyzable input versus misclassification in single-lead wearable ECG},
journal = {Physiological Measurement},
year = {2026},
volume = {47},
number = {11},
pages = {115006},
doi = {10.1088/1361-6579/aebae3}
}
EUVIMED