Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SimuSOE: A Simulated Snoring Dataset for Obstructive Sleep Apnea-Hypopnea Syndrome Evaluation during Wakefulness

Jie Lin, Xiuping Yang, Li Xiao, Xinhong Li, Weiyan Yi, Yuhong Yang, Weiping Tu, Xiong Chen

TL;DR

Experimental results indicate that the simulated snoring signal during wakefulness can serve as an effective feature in OSAHS preliminary screening, and a novel and time-effective snoring collection method is introduced for tackling the above problems.

Abstract

Obstructive Sleep Apnea-Hypopnea Syndrome (OSAHS) is a prevalent chronic breathing disorder caused by upper airway obstruction. Previous studies advanced OSAHS evaluation through machine learning-based systems trained on sleep snoring or speech signal datasets. However, constructing datasets for training a precise and rapid OSAHS evaluation system poses a challenge, since 1) it is time-consuming to collect sleep snores and 2) the speech signal is limited in reflecting upper airway obstruction. In this paper, we propose a new snoring dataset for OSAHS evaluation, named SimuSOE, in which a novel and time-effective snoring collection method is introduced for tackling the above problems. In particular, we adopt simulated snoring which is a type of snore intentionally emitted by patients to replace natural snoring. Experimental results indicate that the simulated snoring signal during wakefulness can serve as an effective feature in OSAHS preliminary screening.

SimuSOE: A Simulated Snoring Dataset for Obstructive Sleep Apnea-Hypopnea Syndrome Evaluation during Wakefulness

TL;DR

Experimental results indicate that the simulated snoring signal during wakefulness can serve as an effective feature in OSAHS preliminary screening, and a novel and time-effective snoring collection method is introduced for tackling the above problems.

Abstract

Obstructive Sleep Apnea-Hypopnea Syndrome (OSAHS) is a prevalent chronic breathing disorder caused by upper airway obstruction. Previous studies advanced OSAHS evaluation through machine learning-based systems trained on sleep snoring or speech signal datasets. However, constructing datasets for training a precise and rapid OSAHS evaluation system poses a challenge, since 1) it is time-consuming to collect sleep snores and 2) the speech signal is limited in reflecting upper airway obstruction. In this paper, we propose a new snoring dataset for OSAHS evaluation, named SimuSOE, in which a novel and time-effective snoring collection method is introduced for tackling the above problems. In particular, we adopt simulated snoring which is a type of snore intentionally emitted by patients to replace natural snoring. Experimental results indicate that the simulated snoring signal during wakefulness can serve as an effective feature in OSAHS preliminary screening.
Paper Structure (14 sections, 2 figures, 4 tables)

This paper contains 14 sections, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Datasets
  3. Data Overview
  4. Data Acquisition
  5. Data Distribution
  6. Experiments
  7. Experimental Setup
  8. Data Split
  9. Evaluation Metrics
  10. Training Details
  11. Results
  12. The classification effects of different screening tasks
  13. Classification effects in different sleeping positions
  14. Conclusions

Figures (2)

  • Figure 1: Recording process for simulated snoring. We record snoring sounds in both supine and lateral positions. When recording the subject's snoring in the lateral position, their entire torso faces to the corresponding side. The microphone is secured 3 cm from the mouth on the patient's face.
  • Figure 2: Confusion matrix comparison of simulated snoring in different sleeping positions on the severe OSAHS screening task (threshold AHI = 30 events/h). Note that we present the confusion matrix of our experiments over three random runs.