Near-boundary double-labeling-based classification: the new standard when evaluating performances of new sleep apnea diagnostic solutions against polysomnography?
(2022) Sleep — Vol. 45, n° 10, p. zsac188 [1-2] (2022)
Files
MartinotJBSleep2022.pdf
Open Access - Adobe PDF
- 363.07 KB
Details
- Authors
Martinot, Jean-Benoît 
Pépin, Jean-Louis
- Abstract
- (en) DEAR EDITOR : Van Pee et al. and the accompanying editorial [2] reported on a very promising methodological approach called near-boundary double-labeling (NBL) method classification for the evaluation of new sleep apnea diagnostic tools. Beyond the usual statistical methods (Bland-Altman, intraclass correlation coefficient, etc.), this complementary analysis led, at least in their peripheral arterial tonometry home sleep apnea test (HSAT) data set (derived from NightOwl technology), to a better agreement with polysomnography (PSG) scored by two independent investigators. This innovative approach allows to assign individuals with apnea hypopnea index (AHI) in predefined near-boundary zones to two different categories of AHI grades of severity. For example, one individual with an AHI close to the threshold of 15 events/hour was labeled with both mild and moderate grades of severity. [...]
- Affiliations
Citations
Martinot, J.-B., Pépin, J.-L., Malhotra, A., & Le-Dong, N.-N. (2022). Near-boundary double-labeling-based classification: the new standard when evaluating performances of new sleep apnea diagnostic solutions against polysomnography? Sleep, 45(10), zsac188 [1-2]. https://doi.org/10.1093/sleep/zsac188 (Original work published 2022)