Microphone and electroglottographic data from dysphonic patients: Type 1, 2 and 3 signals
Microphone and electroglottographic data from dysphonic patients: Type 1, 2 and 3 signals
复制标题
DOI：
10.1016/s0892-1997(98)80045-3
复制
发表时间：
1998-06-01
期刊：
影响因子：
2.2
通讯作者：
Lee, N
中科院分区：
文献类型：
--
作者：
Behrman, A;Agresti, CJ;Lee, N
最近，有人提出，依赖于单个基本周期的可靠提取的统计数据，例如抖动和闪烁，仅对近周期信号有效。本研究探讨了近周期性和非周期性的麦克风和电声门图信号的发生率从202例发音障碍患者。结果发现，约42%是1型（近周期性）;约35%是2型（包含分叉，调制或次谐波结构）;约22%是3型（混沌）。对于40%的最终被评定为类型3的信号，区分类型2和类型3信号非常困难。这是由于明显的混沌段的短暂性，和/或混沌中某些谐波结构的持久性。不管困难，结果表明，在给定的临床人群中，可能有大量的非1型信号的发生。因此，得出结论，信号分型是麦克风和电声门图数据分析的必要步骤。
Recently, it has been suggested that statistics which are dependent upon the reliable extraction of a single fundamental period, such as jitter and shimmer, are valid only for nearly periodic signals. This study explored the incidence of nearly periodic and nonperiodic microphone and electroglottographic signals obtained from 202 dysphonic patients. It was found that approximately 42% were type 1 (nearly periodic); approximately 35% were type 2 (containing bifurcations, modulations or subharmonic structure); and approximately 22% were type 3 (chaotic). Discriminating between type 2 and 3 signals was very difficult for 40% of the signals which were ultimately rated type 3. This was due to the brevity of the apparently chaotic segment, and/or the persistence of some harmonic structure within the chaos. Irrespective of that difficulty, the results suggest that there may be a substantial incidence of nontype 1 signals in a given clinical population It was concluded, therefore, that signal typings is a necessary step in the analyses of microphone and electoglottographic data.