作者: H. Azimi , M. Bouchard , R. A. Goubran , F. Knoefel
DOI: 10.1109/MEMEA.2019.8802214
关键词:
摘要: A method is proposed in this paper for the detection of suspected central apnea events from nocturnal data measured with pressure sensor arrays. Optimized set time and frequency measures computed overlapping segments 9 s are fed to a support vector machine-based classifier identify possible origin segments, i.e., not-apneic or apneic episodes. The decision on sequence successive then used detect complete event. accuracy test data-set overall F-score system found be 94.43% 74.44%, respectively.