Single Sensor Techniques for Sleep Apnea Diagnosis Using Deep Learning

Pathinarupothi, Rahul Krishnan; J, Dhara Prathap; Rangan, Ekanath Srihari; Gopalakrishnan, E. A.; Vinaykumar, R.; Soman, K. P.

doi:10.1109/ichi.2017.37

Cited by 57 publications

(40 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This could possibly indicate that the Apnea-ECG database does not have enough diversity of OSA events, due to the low number of subjects available with the SpO2 signal (8 subjects), thus leading to a high performance of the classifiers. A higher classification performance was also reported by Almazaydeh et al [23], Mostafa et al [25] and Pathinarupothi et al [24] that employed the UCD dataset and the same conclusion can possibly be applied.…”

Section: Discussionsupporting

confidence: 61%

See 1 more Smart Citation

An Oximetry Based Wireless Device for Sleep Apnea Detection

Mendonça

Mostafa

Morgado‐Dias

et al. 2020

Sensors

View full text Add to dashboard Cite

Sleep related disorders can severely disturb the quality of sleep. Among these disorders, obstructive sleep apnea (OSA) is highly prevalent and commonly undiagnosed. Polysomnography is considered to be the gold standard exam for OSA diagnosis. Even though this multi-parametric test provides highly accurate results, it is time consuming, labor-intensive, and expensive. A non-invasive and easy to self-assemble home monitoring device was developed to address these issues. The device can perform the OSA diagnosis at the patient’s home and a specialized technician is not required to supervise the process. An automatic scoring algorithm was developed to examine the blood oxygen saturation signal for a minute-by-minute OSA assessment. It was performed by analyzing statistical and frequency-based features that were fed to a classifier. Afterward, the ratio of the number of minutes classified as OSA to the time in bed in minutes was compared with a threshold for the global (subject-based) OSA diagnosis. The average accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve for the minute-by-minute assessment were, respectively, 88%, 80%, 91%, and 0.86. The subject-based accuracy was 95%. The performance is in the same range as the best state of the art methods for the models based only on the blood oxygen saturation analysis. Therefore, the developed model has the potential to be employed in clinical analysis.

show abstract

Section: Discussionsupporting

confidence: 61%

“…A deep learning approach was proposed by Mostafa et al [22], feeding the raw SpO2 signal to a deep belief network to perform the OSA classification. A similar approach was employed by Pathinarupothi et al [24], using a long short-term memory (LSTM) for classification.…”

Section: Discussionmentioning

confidence: 99%

An Oximetry Based Wireless Device for Sleep Apnea Detection

Mendonça

Mostafa

Morgado‐Dias

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Recent studies that investigated deep learning methods in sleep apnea show improvement over classical machine learning methods [ 46 ]. While the majority of studies considered

signal [ 18 , 19 , 20 , 47 , 48 , 49 , 50 , 51 , 52 , 53 ], a very limited number of them investigated respiration signals. Research efforts with respiratory signals commonly considered multi-channel signal inputs for deep learning.…”

Section: Background and Problem Statementmentioning

confidence: 99%

Deep Recurrent Neural Networks for Automatic Detection of Sleep Apnea from Single Channel Respiration Signals

ElMoaqet

Eid

Glos

et al. 2020

Sensors

View full text Add to dashboard Cite

Sleep apnea is a common sleep disorder that causes repeated breathing interruption during sleep. The performance of automated apnea detection methods based on respiratory signals depend on the signals considered and feature extraction methods. Moreover, feature engineering techniques are highly dependent on the experts’ experience and their prior knowledge about different physiological signals and conditions of the subjects. To overcome these problems, a novel deep recurrent neural network (RNN) framework is developed for automated feature extraction and detection of apnea events from single respiratory channel inputs. Long short-term memory (LSTM) and bidirectional long short-term memory (BiLSTM) are investigated to develop the proposed deep RNN model. The proposed framework is evaluated over three respiration signals: Oronasal thermal airflow (FlowTh), nasal pressure (NPRE), and abdominal respiratory inductance plethysmography (ABD). To demonstrate our results, we use polysomnography (PSG) data of 17 patients with obstructive, central, and mixed apnea events. Our results indicate the effectiveness of the proposed framework in automatic extraction for temporal features and automated detection of apneic events over the different respiratory signals considered in this study. Using a deep BiLSTM-based detection model, the NPRE signal achieved the highest overall detection results with true positive rate (sensitivity) = 90.3%, true negative rate (specificity) = 83.7%, and area under receiver operator characteristic curve = 92.4%. The present results contribute a new deep learning approach for automated detection of sleep apnea events from single channel respiration signals that can potentially serve as a helpful and alternative tool for the traditional PSG method.

show abstract

“…Researchers have proposed many different approaches for SAS detection to overcome the drawbacks of PSG. Contact sensor based detection methods were proposed in [4] – [7] . For example, an off-the-shelf wearable single sensor [4] was used to detect the severity of SAS according to the instantaneous heart rate (IHR) and blood oxygen saturation (SpO2) levels.…”

Section: Introductionmentioning

confidence: 99%

“…Contact sensor based detection methods were proposed in [4] – [7] . For example, an off-the-shelf wearable single sensor [4] was used to detect the severity of SAS according to the instantaneous heart rate (IHR) and blood oxygen saturation (SpO2) levels. A gas sensor array was used to screen SAS, and the performance of the method is susceptible to a number of environmental and metabolic variables [5] .…”

Section: Introductionmentioning

confidence: 99%

Sleep Apnea Syndrome Sensing at C-Band

Yang

Fan

Ren

et al. 2018

IEEE J. Transl. Eng. Health Med.

View full text Add to dashboard Cite

A non-intrusive sleep apnea detection system using a C-Band channel sensing technique is proposed to monitor sleep apnea syndrome in real time. The system utilizes perturbations of RF signals to differentiate between patient’s breathing under normal and sleep apnea conditions. The peak distance calculation is used to obtain the respiratory rates. A comparison of the datasets generated by the proposed method and a wearable sensor is made using a concordance correlation coefficient to establish its accuracy. The results show that the proposed sensing technique exhibits high accuracy and robustness, with more than 80% concordance with the wearable breathing sensor. This method is, therefore, a good candidate for the real-time wireless detection of sleep apnea.

show abstract

Single Sensor Techniques for Sleep Apnea Diagnosis Using Deep Learning

Cited by 57 publications

References 13 publications

An Oximetry Based Wireless Device for Sleep Apnea Detection

An Oximetry Based Wireless Device for Sleep Apnea Detection

Deep Recurrent Neural Networks for Automatic Detection of Sleep Apnea from Single Channel Respiration Signals

Sleep Apnea Syndrome Sensing at C-Band

Contact Info

Product

Resources

About