Scholarlink

Artificial intelligence is a field of computer science that could help pulmonologists diagnose and detect common respiratory disorders such as asthma, COPD, URTI, LRTI, bronchiectasis, pneumonia, and bronchiolitis quickly and reliably. The main objective of this work is to develop an automatic detection method for sibilants to minimize time and optimize diagnostic accuracy. This work is based on the use of deep learning by exploiting a database that contains voice recordings from various categories of individuals. These audios are captured from different parts of the respiratory system, using distinct acquisition modes. We used the convolutional neural network model. The results show that the model performs well with classification results reaching an accuracy of 0.93.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Classification of Respiratory Sounds for Automatic Detection of Sibilants

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Boutaina Rharrab,
  • Larbi Belarbi,
  • Soumia Ziti,
  • Zineb El Ouarrak,
  • Wajih Rhalem

摘要

Artificial intelligence is a field of computer science that could help pulmonologists diagnose and detect common respiratory disorders such as asthma, COPD, URTI, LRTI, bronchiectasis, pneumonia, and bronchiolitis quickly and reliably. The main objective of this work is to develop an automatic detection method for sibilants to minimize time and optimize diagnostic accuracy. This work is based on the use of deep learning by exploiting a database that contains voice recordings from various categories of individuals. These audios are captured from different parts of the respiratory system, using distinct acquisition modes. We used the convolutional neural network model. The results show that the model performs well with classification results reaching an accuracy of 0.93.