Design of embedded real-time system for snoring and OSA detection based on machine learning

Luo, Huaiwen; Li, Heng; Lu, Yun; Lin, Xu; Zhou, Lianyu; Wang, Mingjiang

doi:10.1016/j.measurement.2023.112802

Cited by 6 publications

(4 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In hardware testing, if the model is trained using the original training set and tested using snoring sounds recorded on the hardware platform, the influence of the hardware platform's microphone can lead to discrepancies between the training and test sets. Therefore, before training the model, we adopted the method of our previous work (Luo et al 2023), which used artificial mouth to play snoring sounds. The artificial mouth is a special speaker which can better simulate human vocalization in real-life scene.…”

Section: Evaluations Of Model Hardware Implementationmentioning

confidence: 99%

“…We have previously researched sleep apnea detection models based on snoring (Lin et al 2022, Luo et al 2023. In Luo et al (2023), real-time snoring and OSA detection systems were designed, and a snoring detection algorithm was realized on a hardware platform. In Lin et al (2022), Using snore signals as input, we proposed a sleep apnea detection model based on CNN and LSTM.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Pilot study of contactless sleep apnea detection based on snore signals with hardware implementation

Lin

et al. 2023

Physiol. Meas.

Self Cite

View full text Add to dashboard Cite

Objective. Sleep apnea has a high incidence and is a potentially dangerous disease, and its early detection and diagnosis are challenging. Polysomnography (PSG) is considered the best approach for sleep apnea detection, but it requires cumbersome and complicated operations. Thus, it cannot satisfy the family healthcare needs. Approach. To facilitate the initial detection of sleep apnea in the home environment, we developed a sleep apnea classification model based on snoring and hybrid neural network, and implemented the well trained model in an embedded hardware platform. We used snore signals from 32 patients at Shenzhen People’s Hospital. The Mel-Fbank features were extracted from snore signals to build a sleep apnea classification model based on Bi-LSTM with attention mechanism. Main results. The proposed model classified snore signals into four types: hypopnea, normal condition, obstructive sleep apnea, and central sleep apnea, with 83.52% and 62.31% accuracies, corresponding to the subject-dependence and subject-independence validation, respectively. After pruning and model quantization, at the cost of 0.81% and 0.95% accuracy loss of the subject dependence and subject independence classification, respectively, the number of model parameters and model storage space were reduced by 32.12% and 60.37%, respectively. The model exhibited accuracies of 82.71% and 61.36% based on the subject dependence and subject independence validations, respectively. When the well trained model was successfully porting and running on an STM32 ARM-embedded platform, the model accuracy was 58.85% for the four classifications based on leave-one-subject-out validation. Significance. The proposed sleep apnea detection model can be used in home healthcare for the initial detection of sleep apnea.

show abstract

Section: Evaluations Of Model Hardware Implementationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Pilot study of contactless sleep apnea detection based on snore signals with hardware implementation

Lin

et al. 2023

Physiol. Meas.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Compared with the traditional methods of sound feature extraction, the OSA method based on the deep CNN can automatically extract the information in snoring and learn the representation of the data. This type of method also effectively overcomes the limitations of manually extracting snoring features, such as insufficient feature extraction ability and poor adaptability (Luo et al 2020, Kwon et al 2021, Chen et al 2022, Cheng et al 2022, Kayabekir et al 2022, Luo et al 2023.…”

Section: Introductionmentioning

confidence: 99%

“…known as SNOROSALAB before PSG diagnosis. They used a small part of the experimental samples to propose a complete diagnostic process, and the AHI index obtained by comparing the results with Cohen's kappa criteria had no significant difference Luo et al (2023). trained a multiclassification temporal convolutional network (TCN) to classify nighttime audio as non-snoring, snoring, or OSA-related snoring.…”

mentioning

confidence: 99%

SST: a snore shifted-window transformer method for potential obstructive sleep apnea patient diagnosis

Luo,

Zhao,

Liu

et al. 2024

Physiol. Meas.

View full text Add to dashboard Cite

Objective: Obstructive sleep apnea (OSA) is a high-incidence disease that is seriously harmful and potentially dangerous. The objective of this study was to develop a noncontact sleep audio signal-based method for diagnosing potential obstructive sleep apnea (OSA) patients, aiming to provide a more convenient diagnostic approach compared to the traditional polysomnography (PSG) testing. Approach: The study employed a shifted window transformer model to detect snoring audio signals from whole-night sleep audio. First, a snoring detection model was trained on large-scale audio datasets. Subsequently, the deep feature statistical metrics of the detected snore audio were used to train a random forest classifier for OSA patient diagnosis. Main results:Using a self-collected dataset of 305 potential OSA patients, the proposed snore shifted-window transformer method (SST) achieved an accuracy of 85.9\%, a sensitivity of 85.3\%, and a precision of 85.6\% in OSA patient classification. These values surpassed the state-of-the-art method by 9.7\%, 10.7\%, and 7.9\%, respectively. Significance: The experimental results demonstrated that SST significantly improved the noncontact audio-based OSA diagnosis performance. The study's findings suggest a promising self-diagnosis method for potential OSA patients, potentially reducing the need for invasive and inconvenient diagnostic procedures.

show abstract

CNN-SENet: A Convolutional Neural Network Model for Audio Snoring Detection Based on Channel Attention Mechanism

Mao,

Duan,

Zhang

et al. 2024

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Design of embedded real-time system for snoring and OSA detection based on machine learning

Cited by 6 publications

References 43 publications

Pilot study of contactless sleep apnea detection based on snore signals with hardware implementation

Pilot study of contactless sleep apnea detection based on snore signals with hardware implementation

SST: a snore shifted-window transformer method for potential obstructive sleep apnea patient diagnosis

CNN-SENet: A Convolutional Neural Network Model for Audio Snoring Detection Based on Channel Attention Mechanism

Contact Info

Product

Resources

About