Detection of sleep breathing sound based on artificial neural network analysis

Auditory Property-Based Features and Artificial Neural Network Classifiers for the Automatic Detection of Low-Intensity Snoring/Breathing Episodes

The definitive diagnosis of obstructive sleep apnea syndrome (OSAS) is made using an overnight polysomnography (PSG) test. This test requires that a patient wears multiple measurement sensors during an overnight hospitalization. However, this setup imposes physical constraints and a heavy burden on the patient. Recent studies have reported on another technique for conducting OSAS screening based on snoring/breathing episodes (SBEs) extracted from recorded data acquired by a noncontact microphone. However, SBEs have a high dynamic range and are barely audible at intensities >90 dB. A method is needed to detect SBEs even in low-signal-to-noise-ratio (SNR) environments. Therefore, we developed a method for the automatic detection of low-intensity SBEs using an artificial neural network (ANN). However, when considering its practical use, this method required further improvement in terms of detection accuracy and speed. To accomplish this, we propose in this study a new method to detect low SBEs based on neural activity pattern (NAP)-based cepstral coefficients (NAPCC) and ANN classifiers. Comparison results of the leave-one-out cross-validation demonstrated that our proposed method is superior to previous methods for the classification of SBEs and non-SBEs, even in low-SNR conditions (accuracy: 85.99 ± 5.69% vs. 75.64 ± 18.8%).

Detection of Snore from OSAHS Patients Based on Deep Learning

Obstructive sleep apnea-hypopnea syndrome (OSAHS) is extremely harmful to the human body and may cause neurological dysfunction and endocrine dysfunction, resulting in damage to multiple organs and multiple systems throughout the body and negatively affecting the cardiovascular, kidney, and mental systems. Clinically, doctors usually use standard PSG (Polysomnography) to assist diagnosis. PSG determines whether a person has apnea syndrome with multidimensional data such as brain waves, heart rate, and blood oxygen saturation. In this paper, we have presented a method of recognizing OSAHS, which is convenient for patients to monitor themselves in daily life to avoid delayed treatment. Firstly, we theoretically analyzed the difference between the snoring sounds of normal people and OSAHS patients in the time and frequency domains. Secondly, the snoring sounds related to apnea events and the nonapnea related snoring sounds were classified by deep learning, and then, the severity of OSAHS symptoms had been recognized. In the algorithm proposed in this paper, the snoring data features are extracted through the three feature extraction methods, which are MFCC, LPCC, and LPMFCC. Moreover, we adopted CNN and LSTM for classification. The experimental results show that the MFCC feature extraction method and the LSTM model have the highest accuracy rate which was 87% when it is adopted for binary-classification of snoring data. Moreover, the AHI value of the patient can be obtained by the algorithm system which can determine the severity degree of OSAHS.

Current Status and Future Challenges of Sleep Monitoring Systems: Systematic Review

Background

Sleep is essential for human health. Considerable effort has been put into academic and industrial research and in the development of wireless body area networks for sleep monitoring in terms of nonintrusiveness, portability, and autonomy. With the help of rapid advances in smart sensing and communication technologies, various sleep monitoring systems (hereafter, sleep monitoring systems) have been developed with advantages such as being low cost, accessible, discreet, contactless, unmanned, and suitable for long-term monitoring.

Objective

This paper aims to review current research in sleep monitoring to serve as a reference for researchers and to provide insights for future work. Specific selection criteria were chosen to include articles in which sleep monitoring systems or devices are covered.

Methods

This review investigates the use of various common sensors in the hardware implementation of current sleep monitoring systems as well as the types of parameters collected, their position in the body, the possible description of sleep phases, and the advantages and drawbacks. In addition, the data processing algorithms and software used in different studies on sleep monitoring systems and their results are presented. This review was not only limited to the study of laboratory research but also investigated the various popular commercial products available for sleep monitoring, presenting their characteristics, advantages, and disadvantages. In particular, we categorized existing research on sleep monitoring systems based on how the sensor is used, including the number and type of sensors, and the preferred position in the body. In addition to focusing on a specific system, issues concerning sleep monitoring systems such as privacy, economic, and social impact are also included. Finally, we presented an original sleep monitoring system solution developed in our laboratory.

Results

By retrieving a large number of articles and abstracts, we found that hotspot techniques such as big data, machine learning, artificial intelligence, and data mining have not been widely applied to the sleep monitoring research area. Accelerometers are the most commonly used sensor in sleep monitoring systems. Most commercial sleep monitoring products cannot provide performance evaluation based on gold standard polysomnography.

Conclusions

Combining hotspot techniques such as big data, machine learning, artificial intelligence, and data mining with sleep monitoring may be a promising research approach and will attract more researchers in the future. Balancing user acceptance and monitoring performance is the biggest challenge in sleep monitoring system research.

SnoreNet: Detecting Snore Events from Raw Sound Recordings

Snoring is one of the earliest symptoms of Obstructive Sleep Apnea Hypopnea Syndrome (OSAHS). Snore detection is the first step in developing non-invasive, low-cost, and totally sound-based OSAHS analysis approaches. In this work, we propose a simple yet effective deep neural network, named SnoreNet, for detecting snores from a continuous sound recording. Without manually crafted features, SnoreNet can capture the characteristics of snores. Since snore varies in temporal length, SnoreNet combines output from multiple feature maps to detect snore. In each feature map, SnoreNet uses a set of default bounding box generated by a base length and different scales to match snores. SnoreNet adjusts the box to better locate snores and predicts a score for the presence of snore in each default bounding box. The performance of SnoreNet was evaluated on a newly collected snore pattern classes dataset, which achieves 81.82% average precision (AP).

A Deep Learning Model for Snoring Detection and Vibration Notification Using a Smart Wearable Gadget

Snoring, a form of sleep-disordered breathing, interferes with sleep quality and quantity, both for the person who snores and often for the person who sleeps with the snorer. Poor sleep caused by snoring can create significant physical, mental, and economic problems. A simple and natural solution for snoring is to sleep on the side, instead of sleeping on the back. In this project, a deep learning model for snoring detection is developed and the model is transferred to an embedded system—referred to as the listener module—to automatically detect snoring. A novel wearable gadget is developed to apply a vibration notification on the upper arm until the snorer sleeps on the side. The gadget is rechargeable, and it is wirelessly connected to the listener module using low energy Bluetooth. A smartphone app—connected to the listener module using home Wi-Fi—is developed to log the snoring events with timestamps, and the data can be transferred to a physician for treating and monitoring diseases such as sleep apnea. The snoring detection deep learning model has an accuracy of 96%. A prototype system consisting of the listener module, the wearable gadget, and a smartphone app has been developed and tested successfully.