Scholarlink

<p>A prominent sign of sleep disorders, notably obstructive sleep apnea-hypopnea syndrome (OSAHS) is snoring, leads major health risks if not treated. Even with its widespread nature, snoring is still a difficult to identify accurately and promptly. The signal processing and machine learning algorithms are employed in previous works to deal with this problem, often needs more robust and reliable methods for snoring detection. The proposed study innovatively integrates the convolutional neural networks-Gated Recurrent Unit (CNN-GRU) architecture with IoT to present a novel method for snoring detection and vibration alerting. To accurately classifying the snoring events, this model ensures a utilization of deep learning’s capacity to extracts complex properties from audio data. To deliver the real-time feedback and vibration notifications depends on identified snoring patterns that integrated with smartphones. The study results show a significant result as 99.09% F1-score, accuracy, precision, and recall, that illustrates the proposed method effectiveness in detecting snoring episodes and minimises false signals. Additionally, the integration of loT technology with smartphones, users can easily monitor and receives a notification about snoring condition, enable them to take preventative action against the sleep-related problems. This method ultimately enhances sleep quality and general health outcomes for those people with sleep disorders.</p>

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

Novel CNN-GRU Framework For Snoring Event Prediction - Vibrating Alert

  • 期刊论文

  • 发表日期：

    2025年6月16日

查看全文
  • S. Kiruthika,
  • G. Prakash

摘要

A prominent sign of sleep disorders, notably obstructive sleep apnea-hypopnea syndrome (OSAHS) is snoring, leads major health risks if not treated. Even with its widespread nature, snoring is still a difficult to identify accurately and promptly. The signal processing and machine learning algorithms are employed in previous works to deal with this problem, often needs more robust and reliable methods for snoring detection. The proposed study innovatively integrates the convolutional neural networks-Gated Recurrent Unit (CNN-GRU) architecture with IoT to present a novel method for snoring detection and vibration alerting. To accurately classifying the snoring events, this model ensures a utilization of deep learning’s capacity to extracts complex properties from audio data. To deliver the real-time feedback and vibration notifications depends on identified snoring patterns that integrated with smartphones. The study results show a significant result as 99.09% F1-score, accuracy, precision, and recall, that illustrates the proposed method effectiveness in detecting snoring episodes and minimises false signals. Additionally, the integration of loT technology with smartphones, users can easily monitor and receives a notification about snoring condition, enable them to take preventative action against the sleep-related problems. This method ultimately enhances sleep quality and general health outcomes for those people with sleep disorders.