文献速递耳科与耳神经

基于深度学习框架的听性脑干反应波形自动识别

Automatic Recognition of Auditory Brainstem Response Waveforms Using a Deep Learning-Based Framework.
PubMed 文献 · Otolaryngol Head Neck Surg · 2024-06-01

本研究采用描述性研究设计,收集并分析100名不同年龄和听力水平参与者的纯音测听和听性脑干反应(ABR)数据,分为四组,提取时域和频域特征并结合人口学因素,使用增强的Wide&Deep模型和Light-MLP模型训练ABR波形识别。Light-MLP模型在1至4组的识别准确率分别为97.8%、97.2%、93.8%和92.0%,加权平均95.4%;Wide&Deep模型分别为93.4%、90.8%、92.0%和88.3%,加权平均91.0%。两种模型均表现出优异的自动识别准确率,Wide&Deep模型性能略差,作者认为样本量有限,扩大数据集后其性能可能进一步提升。

要点

  • Light-MLP模型对ABR波形的识别准确率在四组中为97.8%、97.2%、93.8%、92.0%。
  • Wide&Deep模型识别准确率在四组中为93.4%、90.8%、92.0%、88.3%。
  • Light-MLP模型加权平均准确率为95.4%,Wide&Deep模型为91.0%。

核心结论

  • Light-MLP模型和Wide&Deep模型在不同年龄和听力水平的参与者中对ABR波形的自动识别均表现出优异的准确率。
  • Wide&Deep模型的性能略差于Light-MLP模型,作者认为部分原因是样本量有限,扩大数据集后其性能可能进一步提升。

研究设计与样本

  • 描述性研究设计,收集并分析100名参与者的纯音测听和ABR数据。
  • 研究在清华大学长庚医院临床听力中心(中国北京)的三级学术医疗中心进行。
  • 参与者按年龄和听力水平分为四组。

主要发现

  • Light-MLP模型在1至4组的识别准确率分别为97.8%、97.2%、93.8%和92.0%,加权平均准确率为95.4%。
  • Wide&Deep模型在1至4组的识别准确率分别为93.4%、90.8%、92.0%和88.3%,加权平均准确率为91.0%。

局限性

  • Wide&Deep模型的性能略差于Light-MLP模型,作者认为部分原因是样本量有限。
  • 作者预期扩大数据集后Wide&Deep模型的性能可能进一步提升。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVE: Recognition of auditory brainstem response (ABR) waveforms may be challenging, particularly for older individuals or those with hearing loss. This study aimed to investigate deep learning frameworks to improve the automatic recognition of ABR waveforms in participants with varying ages and hearing levels. STUDY DESIGN: The research used a descriptive study design to collect and analyze pure tone audiometry and ABR data from 100 participants. SETTING: The research was conducted at a tertiary academic medical center, specifically at the Clinical Audiology Center of Tsinghua Chang Gung Hospital (Beijing, China). METHODS: Data from 100 participants were collected and categorized into four groups based on age and hearing level. Features from both time-domain and frequency-domain ABR signals were extracted and combined with demographic factors, such as age, sex, pure-tone thresholds, stimulus intensity, and original signal sequences to generate feature vectors. An enhanced Wide&Deep model was utilized, incorporating the Light-multi-layer perceptron (MLP) model to train the recognition of ABR waveforms. The recognition accuracy (ACC) of each model was calculated for the overall data set and each group. RESULTS: The ACC rates of the Light-MLP model were 97.8%, 97.2%, 93.8%, and 92.0% for Groups 1 to 4, respectively, with a weighted average ACC rate of 95.4%. For the Wide&Deep model, the ACC rates were 93.4%, 90.8%, 92.0%, and 88.3% for Groups 1 to 4, respectively, with a weighted average ACC rate of 91.0%. CONCLUSION: Both the Light-MLP model and the Wide&Deep model demonstrated excellent ACC in automatic recognition of ABR waveforms across participants with diverse ages and hearing levels. While the Wide&Deep model's performance was slightly poorer than that of the Light-MLP model, particularly due to the limited sample size, it is anticipated that with an expanded data set, the performance of Wide&Deep model may be further improved.

来源:Otolaryngol Head Neck Surg,查看原文。版权归原作者所有。

引用

  1. Liang S, Xu J, Liu H, et al. Automatic Recognition of Auditory Brainstem Response Waveforms Using a Deep Learning-Based Framework. Otolaryngol Head Neck Surg. 2024. PMID: 38822760. doi:10.1002/ohn.840

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。