Speech Signal and Facial Image Processing for Obstructive Sleep Apnea Assessment

Cheeks appearance as a novel predictor of obstructive sleep apnea: the CASA score study

STUDY OBJECTIVES

Four well-established predictors of obstructive sleep apnea (OSA) risk are body mass index, age, sex, and neck circumference. We have previously reported cheeks appearance as an OSA predictor, which may represent a combination of such predictors in a single, readily available feature. This study sought to answer the question: Is cheeks appearance an OSA risk predictor?

METHODS

This was a prospective cross-sectional diagnostic accuracy study based on STARD (standards for reporting diagnostic accuracy studies). Patients undergoing polysomnography to investigate sleep complaints at a sleep clinic affiliated with a university hospital were assessed using cheeks appearance scored 0-3 for volume and 0-3 for flaccidity to create the Cheeks Appearance for Sleep Apnea (CASA) score ranging from 0 to 6. Appearance was judged by 3 blinded and independent evaluators.

RESULTS

Among 265 patients evaluated, 248 were included. Fifty-seven patients had a CASA score of 0 and 191 had a CASA score between 1 and 6. Polysomnography diagnosed 177 of the individuals with OSA; of these, 167 had an altered CASA score. Sensitivity was 87%, specificity was 82%, positive-predictive value was 94%, negative-predictive value was 66%, and accuracy was 86%.

CONCLUSIONS

Our results suggest that combining volume and flaccidity of cheeks appearance in a single index may constitute a reliable OSA predictor. CASA score is a novel predictor of OSA with internal validity in a sleep laboratory adult population. Our findings support further studies to confirm the external validity of this practical diagnostic tool.

CLINICAL TRIAL REGISTRATION

Registry: ClinicalTrials.gov; Name: Cheeks Appearance as a Novel Predictor of Obstructive Sleep Apnea: The CASA Score Study (CASA); URL: https://clinicaltrials.gov/study/NCT04980586; Identifier: NCT04980586.

CITATION

Prikladnicki A, Gomes E, Sousa LCCR, Gonçalves SC, Martinez D. Cheeks appearance as a novel predictor of obstructive sleep apnea: the CASA score study. J Clin Sleep Med. 2024;20(6):879-885.

Obstructive sleep apnea detection during wakefulness: a comprehensive methodological review

Obstructive sleep apnea (OSA) is a chronic condition affecting up to 1 billion people, globally. Despite this spread, OSA is still thought to be underdiagnosed. Lack of diagnosis is largely attributed to the high cost, resource-intensive, and time-consuming nature of existing diagnostic technologies during sleep. As individuals with OSA do not show many symptoms other than daytime sleepiness, predicting OSA while the individual is awake (wakefulness) is quite challenging. However, research especially in the last decade has shown promising results for quick and accurate methodologies to predict OSA during wakefulness. Furthermore, advances in machine learning algorithms offer new ways to analyze the measured data with more precision. With a widening research outlook, the present review compares methodologies for OSA screening during wakefulness, and recommendations are made for avenues of future research and study designs.

Automatic detection of obstructive sleep apnea based on speech or snoring sounds: a narrative review

Background and Objective

Obstructive sleep apnea (OSA) is a common chronic disorder characterized by repeated breathing pauses during sleep caused by upper airway narrowing or collapse. The gold standard for OSA diagnosis is the polysomnography test, which is time consuming, expensive, and invasive. In recent years, more cost-effective approaches for OSA detection based in predictive value of speech and snoring has emerged. In this paper, we offer a comprehensive summary of current research progress on the applications of speech or snoring sounds for the automatic detection of OSA and discuss the key challenges that need to be overcome for future research into this novel approach.

Methods

PubMed, IEEE Xplore, and Web of Science databases were searched with related keywords. Literature published between 1989 and 2022 examining the potential of using speech or snoring sounds for automated OSA detection was reviewed.

Key Content and Findings

Speech and snoring sounds contain a large amount of information about OSA, and they have been extensively studied in the automatic screening of OSA. By importing features extracted from speech and snoring sounds into artificial intelligence models, clinicians can automatically screen for OSA. Features such as formant, linear prediction cepstral coefficients, mel-frequency cepstral coefficients, and artificial intelligence algorithms including support vector machines, Gaussian mixture model, and hidden Markov models have been extensively studied for the detection of OSA.

Conclusions

Due to the significant advantages of noninvasive, low-cost, and contactless data collection, an automatic approach based on speech or snoring sounds seems to be a promising tool for the detection of OSA.

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

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 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.

Sleep apnea syndrome associated with gonadal hormone imbalance (Review)

Patients with obstructive sleep apnea exhibit an increased risk of developing gonadal disorders. Because a notable number of people worldwide have sleep respiratory and reproductive disorders, it is essential to recognize the association between local upper airway dysfunction and its gonadal effects. Repeated breathing pauses cause sleep fragmentation, disorganization of sleep cycles and stages, sympathetic activation, intermittent hypoxemia and systemic inflammation. Nocturnal intermittent hypoxemia has a direct central effect on neurotransmitters, with disturbances in the normal production of hypothalamic-pituitary hormones. Awakenings and micro-awakenings at the end of apneic episodes produce a central stress responsible for hormonal changes and subsequent endocrine imbalances. The aim of the present study was to investigate the impact of obstructive sleep apnea syndrome (OSAS) on gonadal hormonal homeostasis and its consequences. Recognizing and understanding how local upper airway dysfunction causes gonadal imbalance may facilitate better care for patients with OSAS. Although there may be a direct relationship between sleep-disordered breathing and gonadal function mediated by hormones via the hypothalamic-pituitary-gonadal axis, to date, current therapies have not been effective.

Modified oropharyngeal muscle training and scientific vocalization are effective in treating mild-to-moderate obstructive sleep apnea hypoventilation syndrome in adults

BACKGROUND

Obstructive Sleep Apnea-hypopnea Syndrome (OSAHS) has become a major public health challenge globally. Most patients have a concomitant voice disorder. The existing treatment methods have problems.Aims/Objectives: This study investigates the therapeutic effect of adding scientific vocalization to oropharyngeal muscle training on OSAHS patients.

MATERIAL AND METHODS

A total of 30 patients were selected from September 2020 to October 2022. They underwent overnight polysomnography (PSG) and were identified as having mild to moderate obstructive sleep apnea hypoventilation syndrome. They were then chosen for a three-month period of modified oropharyngeal muscle training combined with scientific vocalization.

RESULTS

Out of the 30 selected patients, 26 patients completed the training. Results showed a significant changes in multiple sleep-related indicators. he clinical outcomes were as follows: 7 cases were cured, 13 cases were effective, and 6 cases were ineffective. The overall effective rate was 76.92%.

CONCLUSIONS AND SIGNIFICANCE

The combination of oropharyngeal muscle training and scientific vocalization for the treatment of mild to moderate OSAHS in adults significantly improves several measures used in the treatment of the condition. The method is easy to learn, effective, safe to use, and affordable. It provides more options for the treatment of OSAHS.

[Digital technology and children's maxillofacial management]

The maxillofacial region has multiple functions such as breathing, language, and facial expressions. Children's maxillofacial development is a complex and long process, which is affected by many factors such as genetics, diseases, bad habits and trauma. Early detection, early diagnosis, and early treatment are important concepts in children's maxillofacial management. Digital technology medicine is an emerging technology based on medical imaging and anatomy that has emerged in recent years. The application of this technology in the field of clinical medicine will undoubtedly bring great benefits to children's maxillofacial management. This article summarizes the research on digital technology in children's maxillofacial management, and focuses on the research on children's malocclusion, children's OSA, cleft lip and palate and other related diseases.

Prediction of obstructive sleep apnea using deep learning in 3D craniofacial reconstruction

Background

Obstructive sleep apnea (OSA) is a common sleep disorder. However, current diagnostic methods are labor-intensive and require professionally trained personnel. We aimed to develop a deep learning model using upper airway computed tomography (CT) to predict OSA and to warn the medical technician if a patient has OSA while the patient is undergoing any head and neck CT scan, even for other diseases.

Methods

A total of 219 patients with OSA [apnea-hypopnea index (AHI) ≥10/h] and 81 controls (AHI <10/h) were enrolled. We reconstructed each patient's CT into 3 types (skeletal structures, external skin structures, and airway structures) and captured reconstructed models in 6 directions (front, back, top, bottom, left profile, and right profile). The 6 images from each patient were imported into the ResNet-18 network to extract features and output the probability of OSA using two fusion methods: Add and Concat. Five-fold cross-validation was used to reduce bias. Finally, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated.

Results

All 18 views with Add as the feature fusion performed better than did the other reconstruction and fusion methods. This gave the best performance for this prediction method with an AUC of 0.882.

Conclusions

We present a model for predicting OSA using upper airway CT and deep learning. The model has satisfactory performance and enables CT to accurately identify patients with moderate to severe OSA.

Is There a Relationship between Voice Quality and Obstructive Sleep Apnea Severity and Cumulative Percentage of Time Spent at Saturations below Ninety Percent: Voice Analysis in Obstructive Sleep Apnea Patients

Background and Objectives: Apnea hypopnea index is the most important criterion in determining the severity of obstructive sleep apnea (OSA), while the percentage of the total number of times which oxygen saturation is measured below 90% during polysomnography (CT90%) is important in determining the severity of hypoxemia. As hypoxemia increases, inflammation will also increase in OSA. Inflammation in the respiratory tract may affect phonation. We aimed to determine the effects of the degree of OSA and CT90% on phonation. Materials and Methods: The patients were between the ages of 18−60 years and were divided into four groups: normal, mild, moderate, and severe OSA. Patients were asked to say the vowels /α:/ and /i:/ for 5 s for voice recording. Maximum phonation time (MPT) was recorded. Using the Praat voice analysis program, Jitter%, Shimmer%, harmonics-to-noise ratio (HNR), and f0 values were obtained. Results: Seventy-two patients were included. Vowel sound /α:/; there was a significant difference for Jitter%, Shimmer%, and HNR measurements between the 1st and the 4th group (p < 0.001, p < 0.001, and p < 0.001, respectively) and a correlation between CT90% and Shimmer% and HNR values (p < 0.001 and p < 0.021, respectively). Vowel sound /i:/; there was a significant difference in f0 values between the 1st group and 2nd and 4th groups (p < 0.028 and p < 0.015, respectively), and for Jitter%, Shimmer%, and HNR measurements between the 1st and 4th group (p < 0.04, p < 0.000, and p < 0.000, respectively), and a correlation between CT90% and Shimmer% and HNR values (p < 0.016 and p < 0.003, respectively). The difference was significant in MPT between the 1st group and 3rd and 4th groups (p < 0.03 and p < 0.003, respectively). Conclusions: Glottic phonation can be affected, especially in patients whose AHI scores are ≥15. Voice quality can decrease as the degree of OSA increases. The increase in CT90% can be associated with the worsening of voice and can be used as a predictor in the evaluation of voice disorders in the future.

A Comprehensive Review: Computational Models for Obstructive Sleep Apnea Detection in Biomedical Applications

Obstructive sleep apnea (OSA) is a sleep disorder characterized by periodic episodes of partial or complete upper airway obstruction caused by narrowing or collapse of the pharyngeal airway despite ongoing breathing efforts during sleep. Fall in the blood oxygen saturation and cortical arousals are prompted by this reduction in the airflow which lasts for at least 10 seconds. Impaired labor performance, debilitated quality of life, excessive daytime sleepiness, high snoring, and tiredness even after a whole night's sleep are the primary symptoms of OSA. In due course, the long-standing contributions of OSA culminate in hypertension, arrhythmia, cerebrovascular disease, and heart failure. The traditional diagnostic approach of OSA is the laboratory-based polysomnography (PSG) overnight sleep study, which is a tedious and labor-intensive process that exaggerates the discomfort to the patient. With the advent of computer-aided diagnosis (CAD), automatic detection of OSA has gained increasing interest among researchers in the area of sleep disorders as it influences both diagnostic and therapeutic decisions. The research literature on sleep apnea published during the last decade has been surveyed, focusing on the varied screening approaches accustomed to identifying OSA events and the developmental knowledge offered by multiple contributors from the software perspective. The current study presents an overview of the pathophysiology of OSA, the detection methods, physiological signals related to OSA, the different preprocessing, feature extraction, feature selection, and classification techniques employed for the detection and classification of OSA. Consequently, the research challenges and research gaps in the diagnosis of OSA are identified, critically analyzed, and presented in the best possible light.

Automated Scoring of Respiratory Events in Sleep with a Single Effort Belt and Deep Neural Networks

OBJECTIVE

Automatic detection and analysis of respiratory events in sleep using a single respiratory effort belt and deep learning.

METHODS

Using 9,656 polysomnography recordings from the Massachusetts General Hospital (MGH), we trained a neural network (WaveNet) to detect obstructive apnea, central apnea, hypopnea and respiratory-effort related arousals. Performance evaluation included event-based analysis and apnea-hypopnea index (AHI) stratification. The model was further evaluated on a public dataset, the Sleep-Heart-Health-Study-1, containing 8,455 polysomnographic recordings.

RESULTS

For binary apnea event detection in the MGH dataset, the neural network obtained a sensitivity of 68%, a specificity of 98%, a precision of 65%, a F1-score of 67%, and an area under the curve for the receiver operating characteristics curve and precision-recall curve of 0.93 and 0.71, respectively. AHI prediction resulted in a mean difference of 0.417.8 and a r² of 0.90. For the multiclass task, we obtained varying performances: 84% of all labeled central apneas were correctly classified, whereas this metric was 51% for obstructive apneas, 40% for respiratory effort related arousals and 23% for hypopneas.

CONCLUSION

Our fully automated method can detect respiratory events and assess the AHI accurately. Differentiation of event types is more difficult and may reflect in part the complexity of human respiratory output and some degree of arbitrariness in the criteria used during manual annotation.

SIGNIFICANCE

The current gold standard of diagnosing sleep-disordered breathing, using polysomnography and manual analysis, is time-consuming, expensive, and only applicable in dedicated clinical environments. Automated analysis using a single effort belt signal overcomes these limitations.

Predicting sleep apnea from three-dimensional face photography

STUDY OBJECTIVES

Craniofacial anatomy is recognized as an important predisposing factor in the pathogenesis of obstructive sleep apnea (OSA). This study used three-dimensional (3D) facial surface analysis of linear and geodesic (shortest line between points over a curved surface) distances to determine the combination of measurements that best predicts presence and severity of OSA.

METHODS

3D face photographs were obtained in 100 adults without OSA (apnea-hypopnea index [AHI] < 5 events/h), 100 with mild OSA (AHI 5 to < 15 events/h), 100 with moderate OSA (AHI 15 to < 30 events/h), and 100 with severe OSA (AHI ≥ 30 events/h). Measurements of linear distances and angles, and geodesic distances were obtained between 24 anatomical landmarks from the 3D photographs. The accuracy with which different combinations of measurements could classify an individual as having OSA or not was assessed using linear discriminant analyses and receiver operating characteristic analyses. These analyses were repeated using different AHI thresholds to define presence of OSA.

RESULTS

Relative to linear measurements, geodesic measurements of craniofacial anatomy improved the ability to identify individuals with and without OSA (classification accuracy 86% and 89% respectively, P < .01). A maximum classification accuracy of 91% was achieved when linear and geodesic measurements were combined into a single predictive algorithm. Accuracy decreased when using AHI thresholds ≥ 10 events/h and ≥ 15 events/h to define OSA although greatest accuracy was always achieved using a combination of linear and geodesic distances.

CONCLUSIONS

This study suggests that 3D photographs of the face have predictive value for OSA and that geodesic measurements enhance this capacity.

Estimation of Apnea-Hypopnea Index using Deep Learning on 3D Craniofacial Scans

Obstructive sleep apnea (OSA) is characterized by decreased breathing events that occur through the night, with severity reported as the apnea-hypopnea index (AHI), which is associated with certain craniofacial features. In this study, we used data from 1366 patients collected as part of Stanford Technology Analytics and Genomics in Sleep (STAGES) across 11 US and Canadian sleep clinics and analyzed 3D craniofacial scans with the goal of predicting AHI, as measured using gold standard nocturnal polysomnography (PSG). First, the algorithm detects pre-specified landmarks on mesh objects and aligns scans in 3D space. Subsequently, 2D images and depth maps are generated by rendering and rotating scans by 45-degree increments. Resulting images were stacked as channels and used as input to multi-view convolutional neural networks, which were trained and validated in a supervised manner to predict AHI values derived from PSGs. The proposed model achieved a mean absolute error of 11.38 events/hour, a Pearson correlation coefficient of 0.4, and accuracy for predicting OSA of 67% using 10-fold cross-validation. The model improved further by adding patient demographics and variables from questionnaires. We also show that the model performed at the level of three sleep medicine specialists, who used clinical experience to predict AHI based on 3D scan displays. Finally, we created topographic displays of the most important facial features used by the model to predict AHI, showing importance of the neck and chin area. The proposed algorithm has potential to serve as an inexpensive and efficient screening tool for individuals with suspected OSA.

Multispectral Video Fusion for Non-Contact Monitoring of Respiratory Rate and Apnea

Continuous monitoring of respiratory activity is desirable in many clinical applications to detect respiratory events. Non-contact monitoring of respiration can be achieved with near- and far-infrared spectrum cameras. However, current technologies are not sufficiently robust to be used in clinical applications. For example, they fail to estimate an accurate respiratory rate (RR) during apnea. We present a novel algorithm based on multispectral data fusion that aims at estimating RR also during apnea. The algorithm independently addresses the RR estimation and apnea detection tasks. Respiratory information is extracted from multiple sources and fed into an RR estimator and an apnea detector whose results are fused into a final respiratory activity estimation. We evaluated the system retrospectively using data from 30 healthy adults who performed diverse controlled breathing tasks while lying supine in a dark room and reproduced central and obstructive apneic events. Combining multiple respiratory information from multispectral cameras improved the root mean square error (RMSE) accuracy of the RR estimation from up to 4.64 monospectral data down to 1.60 breaths/min. The median F1 scores for classifying obstructive (0.75 to 0.86) and central apnea (0.75 to 0.93) also improved. Furthermore, the independent consideration of apnea detection led to a more robust system (RMSE of 4.44 vs. 7.96 breaths/min). Our findings may represent a step towards the use of cameras for vital sign monitoring in medical applications.

Voice disorders in severe obstructive sleep apnea patients and comparison of two acoustic analysis software programs: MDVP and Praat

Objective

The purposes of this study were to explore the effect of obstructive sleep apnea-hypopnea syndrome (OSAHS) on the voice by analyzing the acoustic parameters between patients with OSAHS and those without OSAHS and to compare acoustic analyses performed by two software programs (MDVP and Praat).

Methods

Patients with OSAHS (n = 75) and normal controls (n = 46) were asked to produce a sustained sound of the vowel /i/ and were analyzed with electroglottography (EGG), MDVP, and Praat software. A self-rated scale (Voice Handicap Index, VHI-10) and acoustic parameters were compared.

Results

There were no statistically significant differences in the fundamental frequency (F0), jitter, shimmer, noise/harmonic ratio (NHR), contact quotient perturbation (CQP), or contact index perturbation (CIP) between the patient group and the normal group. The VHI-10 values were significantly increased in patients with OSAHS. The receiver operating characteristic (ROC) analysis suggested that the shimmer obtained from MDVP and Praat possessed relatively high accuracy in differentiating patients with OSAHS from healthy individuals. The results for F0, jitter, shimmer, and NHR were significantly different between MDVP and Praat in OSAHS patients. In normal persons, there was a significant difference in NHR; however, no significant differences were found for F0, jitter, or shimmer between the two software programs. The results demonstrated that high correlations were found between values obtained by both software programs.

Conclusions

Patients with OSAHS were prone to vibration irregularity, incomplete glottal closure, hoarseness, and other vocal problems. The two acoustic software programs present different values of acoustic measures. There was a strong correlation and consistency between the parameters calculated by the two software programs.

[What prospects for the sleep apnea syndrome and connected health?]

Connected health is a growing field and can be viewed from different perspectives, particularly in sleep apnea syndrome. The purpose of this review is to show how all these aspects of connected health are already used in the management of sleep apnea syndrome (SAS) and its comorbidities. First, it can give patients a better understanding and a better assessment of their health. It also facilitates their healthcare by allowing them a greater role in their care pathway. For healthcare providers, connected health tools make it possible to set up new procedures for diagnosing and monitoring ambulatory patients, and for the making of joint decisions by health professionals and patients. Finally, for researchers, e-health generates massive amounts of data, thus facilitating the acquisition of knowledge in real life situations and the development of new methodologies for clinical studies that are faster, less expensive and just as reliable. All these considerations are already applicable in the field of sleep apnea, both for proposed treatments and for comorbidities management and for the patient's involvement in his/her care pathway.

The effect of sleep deprivation on objective and subjective measures of facial appearance

The faces of people who are sleep deprived are perceived by others as looking paler, less healthy and less attractive compared to when well rested. However, there is little research using objective measures to investigate sleep-loss-related changes in facial appearance. We aimed to assess the effects of sleep deprivation on skin colour, eye openness, mouth curvature and periorbital darkness using objective measures, as well as to replicate previous findings for subjective ratings. We also investigated the extent to which these facial features predicted ratings of fatigue by others and could be used to classify the sleep condition of the person. Subjects (n = 181) were randomised to one night of total sleep deprivation or a night of normal sleep (8-9 hr in bed). The following day facial photographs were taken and, in a subset (n = 141), skin colour was measured using spectrophotometry. A separate set of participants (n = 63) later rated the photographs in terms of health, paleness and fatigue. The photographs were also digitally analysed with respect to eye openness, mouth curvature and periorbital darkness. The results showed that neither sleep deprivation nor the subjects' sleepiness was related to differences in any facial variable. Similarly, there was no difference in subjective ratings between the groups. Decreased skin yellowness, less eye openness, downward mouth curvature and periorbital darkness all predicted increased fatigue ratings by others. However, the combination of appearance variables could not be accurately used to classify sleep condition. These findings have implications for both face-to-face and computerised visual assessment of sleep loss and fatigue.

Digital Health and Sleep-Disordered Breathing: A Systematic Review and Meta-Analysis

STUDY OBJECTIVES

Sleep disorders in most individuals remain undiagnosed and without treatment. The use of novel tools and mobile technology has the potential to increase access to diagnosis. The objective of this study was to perform a quantitative and qualitative analysis of the available literature evaluating the accuracy of smartphones and portable devices to screen for sleep-disordered breathing (SDB).

METHODS

A literature review was performed between February 18, 2017 and March 15, 2017. We included studies evaluating adults with SDB symptoms through the use mobile phones and/or portable devices, using standard polysomnography as a comparison. A qualitative evaluation of studies was performed with the QUADAS-2 rating. A bivariate random-effects meta-analysis was used to obtain the estimated sensitivity and specificity of screening SDB for four groups of devices: bed/mattress-based, contactless, contact with three or more sensors, and contact with fewer than three sensors. For each group, we also reported positive predictive values and negative predictive values for mild, moderate, and severe obstructive sleep apnea (OSA) screening.

RESULTS

Of the 22 included studies, 18 were pooled in the meta-analysis. Devices that were bed/mattress-based were found to have the best sensitivity overall (0.921, 95% confidence interval [CI] 0.870, 0.953). The sensitivity of contactless devices to detect mild OSA cases was the highest of all groups (0.976, 95% CI 0.899, 0.995), but provided a high false positive rate (0.487, 95% CI 0.137, 0.851). The remaining groups of devices showed low sensitivity and heterogeneous results.

CONCLUSIONS

This study evidenced the limitations and potential use of portable devices in screening patients for SDB. Additional research should evaluate the accuracy of devices when used at home.

Screening patients for risk of sleep apnea using facial photographs

We investigated using frontal and profile facial photographic images for screening patients for risk of sleep apnea. A 180 image pairs were used from patients who were diagnosed using an attended overnight polysomnogram test into controls (AHI<;10/h) and sleep apnea (AHI≥10/h). A series of 35 landmarks and 71 features motivated by craniofacial structure pertinent to upper airway physiology were identified on the photographs. After reducing the dimension of the feature set using recursive feature selection, the features were processed by a Support Vector Machine (SVM). Classification was performed using linear kernel SVM. The accuracy and area under Receiver Operating Curve (ROC) improved when the number of features reduced from 71 to eight top-ranked features. Further improvement was achieved by adding clinical measurements to the selected features resulting in the accuracy of 80% and the area under ROC of 0.83.

Obstructive Sleep Apnea in Women: Study of Speech and Craniofacial Characteristics

Background

Obstructive sleep apnea (OSA) is a common sleep disorder characterized by frequent cessation of breathing lasting 10 seconds or longer. The diagnosis of OSA is performed through an expensive procedure, which requires an overnight stay at the hospital. This has led to several proposals based on the analysis of patients’ facial images and speech recordings as an attempt to develop simpler and cheaper methods to diagnose OSA.

Objective

The objective of this study was to analyze possible relationships between OSA and speech and facial features on a female population and whether these possible connections may be affected by the specific clinical characteristics in OSA population and, more specifically, to explore how the connection between OSA and speech and facial features can be affected by gender.

Methods

All the subjects are Spanish subjects suspected to suffer from OSA and referred to a sleep disorders unit. Voice recordings and photographs were collected in a supervised but not highly controlled way, trying to test a scenario close to a realistic clinical practice scenario where OSA is assessed using an app running on a mobile device. Furthermore, clinical variables such as weight, height, age, and cervical perimeter, which are usually reported as predictors of OSA, were also gathered.

Acoustic analysis is centered in sustained vowels. Facial analysis consists of a set of local craniofacial features related to OSA, which were extracted from images after detecting facial landmarks by using the active appearance models. To study the probable OSA connection with speech and craniofacial features, correlations among apnea-hypopnea index (AHI), clinical variables, and acoustic and facial measurements were analyzed.

Results

The results obtained for female population indicate mainly weak correlations (r values between .20 and .39). Correlations between AHI, clinical variables, and speech features show the prevalence of formant frequencies over bandwidths, with F2/i/ being the most appropriate formant frequency for OSA prediction in women. Results obtained for male population indicate mainly very weak correlations (r values between .01 and .19). In this case, bandwidths prevail over formant frequencies. Correlations between AHI, clinical variables, and craniofacial measurements are very weak.

Conclusions

In accordance with previous studies, some clinical variables are found to be good predictors of OSA. Besides, strong correlations are found between AHI and some clinical variables with speech and facial features. Regarding speech feature, the results show the prevalence of formant frequency F2/i/ over the rest of features for the female population as OSA predictive feature. Although the correlation reported is weak, this study aims to find some traces that could explain the possible connection between OSA and speech in women. In the case of craniofacial measurements, results evidence that some features that can be used for predicting OSA in male patients are not suitable for testing female population.