Center for Pre-Speech Development and Developmental Disorders, Department of Orthodontics, University of Würzburg, Germany.
Comprehensive Hearing Center (CHC), Department of Otorhinolaryngology, Plastic, Aesthetic and Reconstructive Head and Neck Surgery, University Hospital, University of Würzburg, Germany.
Int J Pediatr Otorhinolaryngol. 2021 May;144:110689. doi: 10.1016/j.ijporl.2021.110689. Epub 2021 Mar 24.
Temporal and fundamental frequency (fo) variations in infant cries provide critical insights into the maturity of vocal control and hearing performances. Earlier research has examined the use of vocalisation properties (in addition to hearing tests) to identify infants at risk of hearing impairment. The aim of this study was to determine whether such an approach could be suitable for neonates.
To investigate this, we recruited 74 healthy neonates within their first week of life as our participants, assigning them to either a group that passed the ABR-based NHS (PG, N = 36) or a group that did not, but were diagnosed as normally hearing in follow-up check at 3 months of life, a so-called false-positive group (NPG, N = 36). Spontaneously uttered cries (N = 2330) were recorded and analysed quantitatively. The duration, minimum, maximum and mean fo, as well as two variability measures (fo range, fo sigma), were calculated for each cry utterance, averaged for individual neonates, and compared between the groups.
A multiple analysis of variance (MANOVA) revealed no significant effects. This confirms that cry features reflecting vocal control do not differ between healthy neonates with normal hearing, irrespective of the outcome of their initial NHS.
Healthy neonates who do not pass the NHS but are normal hearing in the follow-up (false positive cases) have the same cry properties as those with normal hearing who do. This is an essential prerequisite to justify the research strategy of incorporating vocal analysis into NHS to complement ABR measures in identifying hearing-impaired newborns.
婴儿哭声中的时频和基频（fo）变化为发声控制和听力表现的成熟度提供了重要的见解。早期的研究已经检查了使用发声特性（除了听力测试之外）来识别有听力障碍风险的婴儿。本研究的目的是确定这种方法是否适用于新生儿。
为此，我们招募了 74 名健康的新生儿作为参与者，他们在出生后的第一周内，将他们分为通过基于 ABR 的 NHS 的组（PG，N=36）或未通过但在 3 个月的随访检查中被诊断为正常听力的组，即所谓的假阳性组（NPG，N=36）。记录并定量分析自发发声的哭声（N=2330）。计算每个哭声的持续时间、最小、最大和平均 fo，以及两个可变性测量值（fo 范围、fo 西格玛），将每个哭声的测量值平均分配给个体新生儿，并在组间进行比较。
多变量方差分析（MANOVA）未显示出显著的影响。这证实了反映发声控制的哭声特征在具有正常听力的健康新生儿之间没有差异，无论他们最初的 NHS 结果如何。
未通过 NHS 但在随访中听力正常的健康新生儿（假阳性病例）与听力正常的 NHS 通过率高的新生儿具有相同的哭声特征。这是将声音分析纳入 NHS 以补充 ABR 测量来识别听力受损新生儿的研究策略的重要前提。
