Prediction of Oxygen Desaturation by Using Sound Data From a Noncontact Device: A Proof‐of‐Concept Study

Kim, Jeong‐Whun; Shin, Jaeyoung; Lee, Kyogu; Won, Tae‐Bin; Rhee, Chae Seo; Cho, Sung‐Woo

doi:10.1002/lary.29971

Cited by 4 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning, compared with traditional statistical methods, has special advantages on feature recognition and reduce interference of data collection [33] . Kim used contactless device to collect sound data for predicting oxygen desaturation and acquired satis ed results with AUC up to 0.90 [34] . Fernando [35] and Marina Tyan [36] used active appearance model (AAM) to identify facial landmarks and relied on support vector regression (SVR) to estimate the AHI.…”

Section: The Novelty and Bias Of Feature Points Recognitionmentioning

confidence: 99%

The Role of Mandibular Disharmony for Adult Obstructive Sleep Apnea in the Machine-learning Facial Recognition

Chen

Liang

Wang

et al. 2022

Preprint

View full text Add to dashboard Cite

Purpose: The diagnosis of obstructive sleep apnea (OSA) relies on time-consuming and complicated procedures which are not always readily available and may delay the diagnosis. With the widespread use of Artificial Intelligence, quick identification with simple clinical information and image recognition pointing at craniofacial features might be a useful tool for self-helped screening of OSA. Methods: The subjects suspected of OSA receiving sleep examination and frontal photographing were consecutively recruited. Sixty-eight points were labelled with automated identification. An optimized model with facial features and basic clinical information was established and ten-folds cross-validation was performed. Area under the receiver operating characteristic curve (AUC) was calculated to evaluate the model’s performance using sleep monitoring as the reference standard. Results: A total of 653 subjects (77.2% males, 55.3% OSA) were analyzed. CATBOOST was the most suitable algorithm for OSA classification with a sensitivity, specificity, accuracy and AUC of 0.75, 0.66, 0.71 and 0.76 respectively (P<0.05), which was better than STOP-Bang questionnaire, NoSAS scores and Epworth scale. And its advantage was more robust in the prediction of supine sleep apnea with a sensitivity of 0.94. Witnessed apnea by sleep partner was the most powerful variable and followed by body mass index, neck circumference, facial parameters and hypertension. Conclusion: OSA could be identified by a machine-learning derived model with automatic recognition of facial photo for Chinese adults, which may facilitate screening of suspected subjects in a simple and quick manner by mobile application. Clinical trial registration Chinese Clinical Trial Registry: No. ChiCTR-ROC-17011027 (http://chictr.org.cn.)

show abstract

Section: The Novelty and Bias Of Feature Points Recognitionmentioning

confidence: 99%

The Role of Mandibular Disharmony for Adult Obstructive Sleep Apnea in the Machine-learning Facial Recognition

Chen

Liang

Wang

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Apnea-hypopnea index (AHI) and oxygen desaturation index (ODI) are the criteria mostly used in the diagnosis of OSA [1,13]. AHI is represented by the number of apnea and hypopnea events per hour of sleep.…”

Section: Introductionmentioning

confidence: 99%

Is Obstructive Sleep Apnea a Risk Factor for Severe COVID-19?

Zerman¹,

Aydın²

2023

Sleep Med Res

View full text Add to dashboard Cite

Background and Objective Obstructive sleep apnea (OSA) is a breathing disorder during sleep with an obstruction of the upper airway. Previous studies showed OSA as a risk factor for severe COVID-19. It is crucial to determine the risk factors for the severity of COVID-19 infection. We aim to clarify the relationship between COVID-19 severity, and OSA and its degree.Methods Our case-control study included subjects diagnosed with OSA with polysomnography between January 2018 and November 2021, and a control group with a history of COVID-19 infection and without OSA diagnosis. The demographic data, comorbidities, apnea–hypopnea index, and oxygen desaturation index were recorded.Results A total of 217 patients were included in our study. The degree of OSA was a risk factor in the OSA group (p < 0.05); on the other hand, the diagnosis of OSA by itself was not a risk factor. The apnea–hypopnea index and oxygen desaturation index were not related to the severity of COVID-19 infection (p > 0.05). Diabetes mellitus was associated with increased hospitalization in inpatient clinics (p < 0.05) and intensive care unit (ICU) (p < 0.05). Chronic obstructive pulmonary disease was a risk factor for admission to the ICU (p < 0.05).Conclusions Our results showed that moderate-to-severe OSA patients are a vulnerable population to severe COVID-19 infection, although diagnosis of OSA by itself was not a risk factor.

show abstract

“…

We read with interest the report of the study by Kim et al, which focused on predicting oxygen desaturation using sleep breathing sounds with a non-contact device. 1 However, we wanted to share our criticisms of this proof-of-concept study and our suggestions for improving the accuracy of the oxygen desaturation index (ODI) estimation to shed light on future studies.The authors stated that the ODI could be estimated by evaluating the patients' sleep breathing sounds, body mass index, and age. However, it is known that gender, sleeping position, smoking history, and basal arterial oxygen saturation values also affect the ODI value.

…”

mentioning

confidence: 99%

“…We read with interest the report of the study by Kim et al, which focused on predicting oxygen desaturation using sleep breathing sounds with a non-contact device. 1 However, we wanted to share our criticisms of this proof-of-concept study and our suggestions for improving the accuracy of the oxygen desaturation index (ODI) estimation to shed light on future studies.…”

mentioning

confidence: 99%

In reference to Prediction of Oxygen Desaturation by Using Sound Data From a Noncontact Device: A Proof‐of‐Concept Study

Önal

2022

The Laryngoscope

View full text Add to dashboard Cite

We read with interest the report of the study by Kim et al., which focused on predicting oxygen desaturation using sleep breathing sounds with a non-contact device. 1 However, we wanted to share our criticisms of this proof-of-concept study and our suggestions for improving the accuracy of the oxygen desaturation index (ODI) estimation to shed light on future studies.The authors stated that the ODI could be estimated by evaluating the patients' sleep breathing sounds, body mass index, and age. However, it is known that gender, sleeping position, smoking history, and basal arterial oxygen saturation values also affect the ODI value. 2 In addition, differences in upper airway anatomy should have been considered in such a study. Because the upper airway is made up of pharyngeal muscles that are essential for speech and breathing. Factors that cause variations in upper airway anatomy, such as racial differences, change the activity of the pharyngeal muscles that create sleep breathing sounds and may affect the results of studies on this subject. Due to craniofacial differences, the frequency of obstructive sleep apnea may be different among patients with equivalent body mass indexes. 3 On the other hand, the authors did not consider the patients' sleep staging during the study and evaluated only sleep breathing sounds. However, sleep staging is necessary to evaluate sleep and sleep disorders. This is because various physiological changes occur during different sleep stages. Macro sleep stages changes during NREM, REM, or wakefulness cause changes in pharyngeal muscle tone, breathing pattern, and body movements. 4 All these activities are associated with the emergence of different sleep breathing sounds. In addition, increased upper airway resistance during REM sleep causes a decrease in lung volumes and ventilation-perfusion mismatch. This causes oxygen desaturation and may affect ODI results. 2 Another criticism is that the position of the patients was not taken into account. The supine sleeping position causes a significant reduction in the lung volumes and leads to more oxygen desaturation than the lateral decubitus and prone positions. This can affect the ODI results. In addition, the acoustic characteristics of snoring may change with different body positions, and sleep breathing sounds may vary from night to night, even for the same patient. 5 Consequently, a combination of oxygen desaturation and other parameters is more appropriate to increase the diagnostic capability of a sleep monitor. This way, a precise assessment of symptomatic respiratory events can be performed.

show abstract

Prediction of Oxygen Desaturation by Using Sound Data From a Noncontact Device: A Proof‐of‐Concept Study

Cited by 4 publications

References 26 publications

The Role of Mandibular Disharmony for Adult Obstructive Sleep Apnea in the Machine-learning Facial Recognition

The Role of Mandibular Disharmony for Adult Obstructive Sleep Apnea in the Machine-learning Facial Recognition

Is Obstructive Sleep Apnea a Risk Factor for Severe COVID-19?

In reference to Prediction of Oxygen Desaturation by Using Sound Data From a Noncontact Device: A Proof‐of‐Concept Study

Contact Info

Product

Resources

About